{ "text": "キャビンアテンダンス、ドアアウトゥー・ディパーチャー\nキャビンアテンダンス、ドアアウトゥー・ディパーチャー\n皆様、飛行機のドアが閉まりました。\n電子機器は機内モードなど電波の出ない設定に切り替えるか、電源をお切りください。\nLadies and gentlemen, the doors have closed now.\nPlease switch off electronic devices or set them to airplane mode so that radio waves are not emitted.\nThank you.\n各位貴賓，現在飛機機艙門已經關閉，請您將所有電子電池用品關閉或切換為飛航模式，謝謝\n皆様、スターライアンスメンバーANA東京羽田行き、\n852便をご利用いただきましてありがとうございます。\n機長は松井、チーフパーサーは佐藤でございます。\nご用の際には遠慮なく客室乗務員にお知らせください。\nこの便は、スターライアンスパートナーとのコードシェア便でございます。\n前方のスクリーンで非常用設備をご案内いたします。\nGood afternoon ladies and gentlemen, welcome on board ANA, a member of Star Alliance flight 852 to Tokyo.\nThe captain in command of this flight is Masanori Matsui and I'm chief officer Kaori Sato.\nPlease let us know if you need assistance and we hope you'll enjoy the flight with us.\nThis is a codeshare flight with our Star Alliance partner.\nWe'll now explain the safety features of this aircraft.\nWe ask you for a moment of your attention. Thank you.\n各位貴賓午安\n歡迎搭乘星空聯盟成員ANA飛往東京的NH-852次班機\n本班機是ANA與長榮航空公司的代碼共用班機\n本班機的機長是松井\n此一,座舱長是Sato\n畢信蓮,本班機上共配有三名台機組員為您服務\n若您有任何需要,歡迎隨時用中文與我們聯繫\n現在為您播放機上的緊急安全逃生須知\n請各位貴賓仔細收看,謝謝\n[聴不清／待確認]\n歡迎來到A N A\n請介紹機內的安全\n我們請您注意安全的指示\n請將行李放入上方的物入\n或是坐下的行李架\n[聴不清／待確認]\n脱出の妨げとなるため通路や非常口付近には手荷物を置かないでください\n離陸着陸の際は座席とテーブルを元の位置にお戻しください\nReturn your seat and table to the original position during takeoff and landing.\nベルト着用サイン点灯中並びにお座りの間は常にシートベルトをお締めください\n外す場合は金具の片方を持ち上げてください\nこの便は全席禁煙です\n化粧室でのタバコは航空法で禁止されています\n携帯のタバコはスモーキングです\n携帯電話などの電波を発する状態にある機器はご使用になれません\n電波を発しない設定に切り替えていただくか電源をお切りください\n次に、非常用設備をご案内します\n酸素が必要な状態になると酸素マスクが降りてきます\nマスクを強く引き寄せ鼻と口に当てて呼吸してください\nゴム紐は頭にかけ長さを調整してください\n緊急の場合は乗務員の指示に従い衝撃に備える姿勢をとってください\n非常口は機体の前方、中央、後方にあります\nお近くの非常口をお確かめください\n脱出の際には経路と非常口を示す明かりが点灯します\n乗務員の指示に従い行動してください\n手荷物は一切持たずハイヒールは脱いでください\nスライドを使用する際は両腕を前に伸ばし上体を起こして滑ります\n援助する場合はスライドの横に立ち体を引き起こしてください\n立ち上がりすぐに飛行機から離れてください\n行きましょう。\n救命衣は座席の下または横にあります\n袋から取り出し背当て布を後ろにしてかぶります\n両側の金具を前の留め金にかけ左右に引いて締めます\n非常口または翼の上で前の引き手を強く引いて膨らませてください\n膨らみが足りない場合は両側のゴム管を吹いてください\n座席ポケットの安全のしおりをご覧ください\nIf you have any questions, please contact the cabin attendant.\n皆様、入陸いたします。シートベルトをもう一度お確かめください。\nLadies and gentlemen, we are now ready for take off. Please make sure that your seat belt is security fastened. Thank you.\n各位貴賓,飛機即將起飛,請您再次確認是否確實繫好安全帶,謝謝\n皆さま、羽田空港到着時刻は3月14日午後5時30分の予定でございます。\n松山と東京の間には1時間の時差があり、ただいまの東京の時刻は午後2時40分でございます。\n毛布をご使用の際は、毛布の上からシートベルトをお締めください。\n液体類の機内持ち込み制限についてご案内いたします。\n出国までに購入された一容器当たり100mlを超える液体、ジェル、クリーム、エアゾール類は、\n国際線への乗り継ぎ便及びお帰りの便の客室への持ち込みはできませんので、ご注意ください。\n発熱・腹痛などの症状があるお客様は、客室乗務員にお知らせください。\nベルト着用サインは消えましたが、急な揺れに備え、\nお座りの間は常にシートベルトをお締めください。\nLadies and gentlemen, the captain has turned off the seatbelt sign. However, please keep your seatbelt fastened whenever you are seated.\nWe expect to arrive at Tokyo International Airport.\nat 5 30 p.m. March 14th\nThe time difference between Songshan and Tokyo is 1 hour.\nSo local time in Tokyo is 2.42pm.\n[聴不清／待確認]\nMay we be reminded that liquids, gels, creams, and aerosols exceeding 100 ml capacity purchased before departure are not allowed on board yet connecting into national flight or return flight.\nPassengers, if you have any symptoms such as fever or stomach ache, please inform a cabin attendant. Thank you.\n各位贵宾,本班機预计在当地时间下午5点30分抵达东京羽田国际机场\n现在当地的时间为下午2点42分\n台北和东京之间的时差为1小时\n提醒您,您在旅途中购买的物品\n如果有单个容器超过100毫升的液体\n發膠、乳液等、噴霧機等物品\n在轉乘其他航班及回國時不得帶上飛機\n如果有疑问请与空服人员联系\n若您有發燒、腹痛等症狀的旅客\n也请与空服人员联系,谢谢\n虽然现在安全带指示灯已经熄灭\n但是由于飞行中可能突然遭遇乱流\n请您在就座时仍确实系好安全带\n奥そろぎのところお連れいたします。渡辺基長です。\n左手、ややこうなりますが、桜島がうっすらと水蒸気が上がっているのをご覧いただけます。\n左手下宮古の城址、九州の南端を飛行続けております\nまた、間もなくしますと左手、皆様にも現在やや前方\n熊本の山をご覧いただけるかと思います\n本日は上空非常に不安定な\n気流によりまして通常よりも低めの高度を飛行続けております\n現在23000フィート上空7000メートル上空を飛行続けています\nこの先所々気流の影響を受けまして軽い揺れがあろうかと思います\nできる限り安定した気流を選択しながら飛行を続けてまいりますが\n揺れましても安全上心配ありませんのでどうぞご安心ください\n飛行時間はあと1時間10分ほどで羽田空港に着陸いたします\n天候は雨気温は7度でございます少し冷え込んでいるようでございます\nゲート到着は定刻の17時30分から\nありがとうございました\n15分ほど早く\n本日は17時15分の到着予定しております\nごゆっくりと過ごしてください\nご搭乗を御礼申し上げます\nありがとうございました\nThis is captain speaking. We are cruising 23,000, it's about 7,000 meters and ground speed at 570 knots. And one for Tokyo. I expect a nice ride. Another 30 minutes, and I expect the light or bumpier when we start to descend for Tokyo. So when you have a seat, please fasten your seat belt.\nThe Tokyo's weather is raining condition and 7 degrees Celsius.\nThe Tokyo's weather is raining condition and seven degrees Celsius. Landing time will be at five o' five. A little gate in the five fifteen local time, in fifteen minutes before the schedule. Once again, thank you for voting.\nOnce again, thank you for voting.\n你好，我是志全。感谢乘坐本次航班\n我们乘坐的是到达，祝您旅途愉快，谢谢。\n着陸まではおよそ1時間ほどの飛行でございます。\n化粧室はお早めにご使用ください\n日本入国書類のご記入はお済みでしょうか\nどうぞもう一度お確かめください\n到着地の天候は雨、気温は7度でございます。\n国際線ターミナル111番ゲートに到着いたします。\nLadies and gentlemen, we'll be landing in about one hour.\nPassengers who wish to use the lavatory are kindly asked to do so at this time.\nWe'll be arriving at the gate number 111.\nThe weather is raining and temperature is 7 degrees Celsius. Thank you.\n各位贵宾,飞机预计将在一个小时后抵达东京羽田国际机场\n当地的天气为雨天,气温为7度\n请您再次确认日本的入境资料是否已经填写好\n持有它国护照的旅客必须要填写入国登记卡以及海关申报单\n有需要的貴賓請與空服人員索取\n收到機長指示,稍後飛機將準備下降\n想要使用洗手間的貴賓,請您盡早使用洗手間\n當安全帶指示燈亮起時,請您儘快回到座位\n並確實系好安全帶,謝謝\n皆様、着陸準備のため、免税品販売を終了いたします。\nお買い上げありがとうございました。\nLadies and gentlemen, we have just closed the intermission. Thank you for your purchase.\n各位贵宾我们现在结束免税商品的销售\n感谢你的购买\n皆様、シートベルトは緩みのないようしっかりお締めください。\nLadies and gentlemen, please make sure that your seat belt is securely fastened.\nThank you.\n皆様、ベルト着用サインが点灯いたしました。揺れることが予想されております。\nステートベルトは緩みのないよう、しっかりお締めください。\nLadies and gentlemen, the captain has turned on the seatbelt sign.\nPlease return to your seat and make sure your seat belt is securely fastened. Thank you.\n各位貴賓，現在安全帶指示燈已經亮起，請您確實系好安全帶，並且請勿使用洗手間，謝謝。\n请您确实系好安全带\n并且请勿使用洗手间,谢谢\n皆様、ベルト着用サインは消えましたが、\nまもなく着陸体制に入ります。\nご用のないお客様は、シートベルトをお締めのままお過ごしください。\n化粧室ご使用ご希望のお客様は、ベルト着用サインが消えている間にご使用ください。\nLadies and gentlemen, the captain has turned off the seatbelt sign. However, please keep your seatbelt fastened whenever you're seated. Thank you.\n各位貴賓，雖然現在安全帶指示燈已經熄滅，但是由於飛行中可能突然遭遇亂流，建議在著座時仍確實系好安全帶。\n但是由於飛行中可能突然遭遇亂流\n建議在著座時仍確實系好安全帶\n之後我們將準備降落\n如需要使用洗手間的貴賓請盡快使用,謝謝\n皆様、シートベルトは緩みのないようしっかりお締めください。\nLadies and gentlemen, please make sure that your seatbelt is securely fastened.\nThank you.\n皆様、着陸体制に入りました。\nシートベルトを緩みのないようしっかりお締めください。\n座席のリクライニングとテーブル、フットレスト、シートテレビ、コントローラーを元の位置にお戻しください。\n手荷物は上の物入れ、エコノミークラスは前の座席の下にお入れください。\n化粧室のご使用はお控えください。\nLadies and gentlemen, we are now making our final approach.\nPlease make sure that your seat belt is securely fastened and return your seat, table,\nfootrest, seat TV and controllers to the original position.\nPlease refrain from using the lavatories. Thank you.\n各位贵宾,飞机现在准备降落\n请您确实系好安全带\n請將椅背豎直\n餐桌、腳踏板、座位電視、遙控器等收回原位\n手提行李请放在前方座位的下面或上方的行李架\n皆様、羽田空港に着陸いたしました。\nただいまの時刻は3月14日\n午後5時8分でございます\nベルト着用サインが消えるまで、お座りのままお待ちください。\n物入れを開けたときに手荷物が滑り出る恐れがありますのでお気を付けください\nこの先、通信用の電波を発する電子機器もご使用になれますが、\n携帯電話での通話は周りのお客様のご迷惑になりますので、お控えください。\n果物、野菜、植物、肉類は、\n税関検査の前に、検疫同植物カウンターでの検査が必要です。\n海外で牛や豚、鳥などの家畜にいる場所にいたお客様は\n圏域動植物カウンターにお立ち寄りください\nLadies and gentlemen, we have landed at Tokyo International Airport.\nThe local time in Tokyo is 10 minutes past 5.\nMarch 14th. Please remain seated until the captain turns off the seatbelt sign.\nTake care when opening the overhead compartment.\nFrom now on, you can use electronic devices that emit radio waves.\nHowever, please refrain from talking on cell phones out of consideration for other passengers.\nPassengers carrying fresh fruit, vegetables, plant and meat products must proceed to the\nplant and animal quarantine counter before going to customs.\nIf you have visited any overseas livestock facilities or may come into contact with livestock in Japan, please proceed to the plant and animal quarantine counter.\nThank you for flying with ANA and Star Alliance partner. We look forward to serving you again.\n皆さま、今日もANAとスターライアンスパートナーとのコードシェア便をご利用いただきましてありがとうございました。\n次のご利用をお待ちしております\n飛行機は間もなく停止いたしますが、安全のためベルト着用サインが消えるまでお座りのままお待ちください。\nLadies and gentlemen, the aircraft will come to stop in a moment.\nFor your own safety, please remain seated until the captain turns off the seatbelt sign.\nThank you.\n皆様、機内に忘れ物のないよう、上の物入れ、座席ポケットをお確かめください。\n外は冷たい雨が降っております。どうぞ気をつけてお帰りください。\nありがとうございました。\nLadies and gentlemen, please make sure to take all your belongings with you when you\ndepart. Thank you.\nThank you.", "language": "ja", "duration": 1382.016875, "segments": [ { "start": 2.1, "end": 4.58, "text": "キャビンアテンダンス、ドアアウトゥー・ディパーチャー", "raw_asr_text": "キャビンアテンダンスドアアウトゥー・ディパーチャー", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 0, "words": [], "asr_metrics": { "avg_logprob": -0.17755292963098596, "compression_ratio": 1.7377049180327868, "no_speech_prob": 7.893489681332078e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "キャビンアテンダンスドアアウトゥー・ディパーチャーキャビンアテンダンスドアアウトゥー・ディパーチャー皆様、飛行機のドアが閉まりました。電子機器は機内モードなど電波の出ない設定に切り替えるか、電源をお切りください。" } ], "id": 0, "review_annotation": "待確認", "row_id": 0, "bounded_row_id": 0, "parent_row_ids": [ 0 ], "parent_row_id": 0, "packet_id": 0, "LID_evidence": [ { "region_id": 0, "probabilities": { "ja": 0.820589487751325, "zh": 0.004170929018679696, "en": 0.0892083525735264 }, "status": "STABLE" } ], "VAD_region_ids": [ 0 ], "language_confidence": 0.820589487751325, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1.3, "packet_end": 19.0, "raw_text": "キャビンアテンダンスドアアウトゥー・ディパーチャー", "selected_text": "キャビンアテンダンス、ドアアウトゥー・ディパーチャー", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 0 ], "packet_id": 0, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "キャビンアテンダンスドアアウトゥー・ディパーチャー", "after": "キャビンアテンダンス、ドアアウトゥー・ディパーチャー", "classification": "BEST_SUPPORTED_WITH_REVIEW", "reason": "ASR 'アウトゥー・ディパーチャー' phonetically matches 'out' and 'departure' in standard Japanese announcements." } ], "language": "ja", "packet_ids": [ 0 ], "index": 0, "independent_ASR": { "text": "キャビンアテンダントドアズフォアディパーチャー、キャビンアテンダントドアズフォアディパーチャー。皆様、飛行機のドアが閉まりました。電子機器は機内モードなど電波の出ない設定に切り替えるか電源をお切りください。", "chunks": [ { "text": "キャビンアテンダントドアズフォアディパーチャー、キャビンアテンダントドアズフォアディパーチャー。皆様、飛行機のドアが閉まりました。電子機器は機内モードなど電波の出ない設定に切り替えるか電源をお切りください。", "start": 0.0, "end": 17.7, "chunk_index": 0, "language": "Japanese", "finish_reason": "eos", "truncated": false, "generated_tokens": 76, "max_new_tokens": 245 } ], "detected_language": "Japanese", "segments": [ { "start": 1.3, "end": 19.0, "text": "キャビンアテンダントドアズフォアディパーチャー、キャビンアテンダントドアズフォアディパーチャー。皆様、飛行機のドアが閉まりました。電子機器は機内モードなど電波の出ない設定に切り替えるか電源をお切りください。" } ], "start": 1.3, "end": 19.0, "engine": "mlx-qwen3-asr", "language": "ja" }, "source_text": "キャビンアテンダンス、ドアアウトゥー・ディパーチャー", "review_status": "BEST_SUPPORTED_WITH_REVIEW", "classification": "BEST_SUPPORTED_WITH_REVIEW", "confidence": 0.85, "correction_reason": "ASR 'アウトゥー・ディパーチャー' phonetically matches 'out' and 'departure' in standard Japanese announcements.", "raw_word_timestamps": [ { "word": "キ", "start": 2.0600000000000005, "end": 2.34, "probability": 0.81787109375 }, { "word": "ャ", "start": 2.34, "end": 2.3600000000000003, "probability": 0.99951171875 }, { "word": "ビ", "start": 2.3600000000000003, "end": 2.46, "probability": 0.998046875 }, { "word": "ン", "start": 2.46, "end": 2.52, "probability": 0.98291015625 }, { "word": "ア", "start": 2.52, "end": 2.62, "probability": 0.85986328125 }, { "word": "テ", "start": 2.62, "end": 2.74, "probability": 0.99658203125 }, { "word": "ンダ", "start": 2.74, "end": 2.92, "probability": 0.98876953125 }, { "word": "ン", "start": 2.92, "end": 2.98, "probability": 0.452392578125 }, { "word": "ス", "start": 2.98, "end": 3.08, "probability": 0.82470703125 }, { "word": "ド", "start": 3.08, "end": 3.2, "probability": 0.6591796875 }, { "word": "ア", "start": 3.2, "end": 3.34, "probability": 0.99853515625 }, { "word": "ア", "start": 3.34, "end": 3.46, "probability": 0.5185546875 }, { "word": "ウ", "start": 3.46, "end": 3.5, "probability": 0.83154296875 }, { "word": "ト", "start": 3.5, "end": 3.5200000000000005, "probability": 0.7060546875 }, { "word": "ゥ", "start": 3.5200000000000005, "end": 3.6399999999999997, "probability": 0.6298828125 }, { "word": "ー", "start": 3.6399999999999997, "end": 3.74, "probability": 0.273193359375 }, { "word": "・", "start": 3.74, "end": 3.8, "probability": 0.63623046875 }, { "word": "デ", "start": 3.8, "end": 3.88, "probability": 0.8359375 }, { "word": "ィ", "start": 3.88, "end": 3.9400000000000004, "probability": 0.73193359375 }, { "word": "パ", "start": 3.9400000000000004, "end": 4.04, "probability": 0.9794921875 }, { "word": "ーチ", "start": 4.04, "end": 4.3, "probability": 0.99560546875 }, { "word": "ャ", "start": 4.3, "end": 4.4, "probability": 0.99853515625 }, { "word": "ー", "start": 4.4, "end": 4.58, "probability": 0.9970703125 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 4.58, "end": 7.62, "text": "キャビンアテンダンス、ドアアウトゥー・ディパーチャー", "raw_asr_text": "キャビンアテンダンスドアアウトゥー・ディパーチャー", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 0, "words": [ { "word": "キ", "start": 4.58, "end": 4.96, "probability": 0.271728515625 }, { "word": "ャ", "start": 4.96, "end": 5.0200000000000005, "probability": 1.0 }, { "word": "ビ", "start": 5.0200000000000005, "end": 5.12, "probability": 1.0 }, { "word": "ン", "start": 5.12, "end": 5.22, "probability": 1.0 }, { "word": "ア", "start": 5.22, "end": 5.34, "probability": 0.99853515625 }, { "word": "テ", "start": 5.34, "end": 5.4799999999999995, "probability": 0.99951171875 }, { "word": "ンダ", "start": 5.4799999999999995, "end": 5.72, "probability": 0.9990234375 }, { "word": "ン", "start": 5.72, "end": 5.819999999999999, "probability": 0.99658203125 }, { "word": "ス", "start": 5.819999999999999, "end": 5.9399999999999995, "probability": 0.9990234375 }, { "word": "ド", "start": 5.9399999999999995, "end": 6.2, "probability": 0.984375 }, { "word": "ア", "start": 6.2, "end": 6.359999999999999, "probability": 1.0 }, { "word": "ア", "start": 6.359999999999999, "end": 6.4799999999999995, "probability": 0.99609375 }, { "word": "ウ", "start": 6.4799999999999995, "end": 6.5, "probability": 0.9990234375 }, { "word": "ト", "start": 6.5, "end": 6.54, "probability": 0.99951171875 }, { "word": "ゥ", "start": 6.54, "end": 6.68, "probability": 1.0 }, { "word": "ー", "start": 6.68, "end": 6.74, "probability": 0.99560546875 }, { "word": "・", "start": 6.74, "end": 6.8, "probability": 0.9990234375 }, { "word": "デ", "start": 6.8, "end": 6.88, "probability": 0.99560546875 }, { "word": "ィ", "start": 6.88, "end": 6.9399999999999995, "probability": 0.99951171875 }, { "word": "パ", "start": 6.9399999999999995, "end": 7.02, "probability": 0.99951171875 }, { "word": "ーチ", "start": 7.02, "end": 7.26, "probability": 0.99951171875 }, { "word": "ャ", "start": 7.26, "end": 7.34, "probability": 1.0 }, { "word": "ー", "start": 7.34, "end": 7.62, "probability": 1.0 } ], "asr_metrics": { "avg_logprob": -0.17755292963098596, "compression_ratio": 1.7377049180327868, "no_speech_prob": 7.893489681332078e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "キャビンアテンダンスドアアウトゥー・ディパーチャーキャビンアテンダンスドアアウトゥー・ディパーチャー皆様、飛行機のドアが閉まりました。電子機器は機内モードなど電波の出ない設定に切り替えるか、電源をお切りください。" } ], "id": 1, "review_annotation": "待確認", "row_id": 1, "bounded_row_id": 1, "parent_row_ids": [ 1 ], "parent_row_id": 1, "packet_id": 0, "LID_evidence": [ { "region_id": 0, "probabilities": { "ja": 0.820589487751325, "zh": 0.004170929018679696, "en": 0.0892083525735264 }, "status": "STABLE" } ], "VAD_region_ids": [ 0 ], "language_confidence": 0.820589487751325, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1.3, "packet_end": 19.0, "raw_text": "キャビンアテンダンスドアアウトゥー・ディパーチャー", "selected_text": "キャビンアテンダンス、ドアアウトゥー・ディパーチャー", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 1 ], "packet_id": 0, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "キャビンアテンダンスドアアウトゥー・ディパーチャー", "after": "キャビンアテンダンス、ドアアウトゥー・ディパーチャー", "classification": "BEST_SUPPORTED_WITH_REVIEW", "reason": "Repeated phrase; same phonetic reconstruction as row 0 applies." } ], "language": "ja", "packet_ids": [ 0 ], "index": 1, "independent_ASR": { "text": "キャビンアテンダントドアズフォアディパーチャー、キャビンアテンダントドアズフォアディパーチャー。皆様、飛行機のドアが閉まりました。電子機器は機内モードなど電波の出ない設定に切り替えるか電源をお切りください。", "chunks": [ { "text": "キャビンアテンダントドアズフォアディパーチャー、キャビンアテンダントドアズフォアディパーチャー。皆様、飛行機のドアが閉まりました。電子機器は機内モードなど電波の出ない設定に切り替えるか電源をお切りください。", "start": 0.0, "end": 17.7, "chunk_index": 0, "language": "Japanese", "finish_reason": "eos", "truncated": false, "generated_tokens": 76, "max_new_tokens": 245 } ], "detected_language": "Japanese", "segments": [ { "start": 1.3, "end": 19.0, "text": "キャビンアテンダントドアズフォアディパーチャー、キャビンアテンダントドアズフォアディパーチャー。皆様、飛行機のドアが閉まりました。電子機器は機内モードなど電波の出ない設定に切り替えるか電源をお切りください。" } ], "start": 1.3, "end": 19.0, "engine": "mlx-qwen3-asr", "language": "ja" }, "source_text": "キャビンアテンダンス、ドアアウトゥー・ディパーチャー", "review_status": "BEST_SUPPORTED_WITH_REVIEW", "classification": "BEST_SUPPORTED_WITH_REVIEW", "confidence": 0.85, "correction_reason": "Repeated phrase; same phonetic reconstruction as row 0 applies." }, { "start": 8.38, "end": 10.58, "text": "皆様、飛行機のドアが閉まりました。", "raw_asr_text": "皆様、飛行機のドアが閉まりました。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 0, "words": [ { "word": "皆", "start": 8.38, "end": 8.66, "probability": 0.97705078125 }, { "word": "様、", "start": 8.66, "end": 8.94, "probability": 0.57373046875 }, { "word": "飛", "start": 9.040000000000001, "end": 9.14, "probability": 0.99853515625 }, { "word": "行", "start": 9.14, "end": 9.280000000000001, "probability": 0.99951171875 }, { "word": "機", "start": 9.280000000000001, "end": 9.42, "probability": 0.99951171875 }, { "word": "の", "start": 9.42, "end": 9.540000000000001, "probability": 1.0 }, { "word": "ド", "start": 9.540000000000001, "end": 9.66, "probability": 0.99951171875 }, { "word": "ア", "start": 9.66, "end": 9.760000000000002, "probability": 1.0 }, { "word": "が", "start": 9.760000000000002, "end": 9.9, "probability": 1.0 }, { "word": "閉", "start": 9.9, "end": 10.08, "probability": 0.99951171875 }, { "word": "ま", "start": 10.08, "end": 10.14, "probability": 0.99267578125 }, { "word": "りました。", "start": 10.14, "end": 10.58, "probability": 0.99951171875 } ], "asr_metrics": { "avg_logprob": -0.17755292963098596, "compression_ratio": 1.7377049180327868, "no_speech_prob": 7.893489681332078e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "キャビンアテンダンスドアアウトゥー・ディパーチャーキャビンアテンダンスドアアウトゥー・ディパーチャー皆様、飛行機のドアが閉まりました。電子機器は機内モードなど電波の出ない設定に切り替えるか、電源をお切りください。" } ], "id": 2, "review_annotation": "", "row_id": 2, "bounded_row_id": 2, "parent_row_ids": [ 2 ], "parent_row_id": 2, "packet_id": 0, "LID_evidence": [ { "region_id": 1, "probabilities": { "ja": 0.9779892414808273, "zh": 0.0009155690495390445, "en": 0.012302618621712705 }, "status": "STABLE" } ], "VAD_region_ids": [ 1 ], "language_confidence": 0.9779892414808273, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1.3, "packet_end": 19.0, "raw_text": "皆様、飛行機のドアが閉まりました。", "selected_text": "皆様、飛行機のドアが閉まりました。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 2 ], "packet_id": 0, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "皆様、飛行機のドアが閉まりました。", "after": "皆様、飛行機のドアが閉まりました。", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard Japanese announcement text, clearly supported by context and English parallel." } ], "language": "ja", "packet_ids": [ 0 ], "index": 2, "source_text": "皆様、飛行機のドアが閉まりました。", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard Japanese announcement text, clearly supported by context and English parallel." }, { "start": 10.940000000000001, "end": 17.68, "text": "電子機器は機内モードなど電波の出ない設定に切り替えるか、電源をお切りください。", "raw_asr_text": "電子機器は機内モードなど電波の出ない設定に切り替えるか、電源をお切りください。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 0, "words": [ { "word": "電", "start": 10.940000000000001, "end": 11.22, "probability": 0.9794921875 }, { "word": "子", "start": 11.22, "end": 11.38, "probability": 0.99365234375 }, { "word": "機", "start": 11.38, "end": 11.5, "probability": 0.9990234375 }, { "word": "器", "start": 11.5, "end": 11.66, "probability": 0.99951171875 }, { "word": "は", "start": 11.66, "end": 11.88, "probability": 0.99853515625 }, { "word": "機", "start": 11.88, "end": 12.600000000000001, "probability": 0.65185546875 }, { "word": "内", "start": 12.600000000000001, "end": 12.8, "probability": 0.99951171875 }, { "word": "モ", "start": 12.8, "end": 13.020000000000001, "probability": 0.99951171875 }, { "word": "ード", "start": 13.020000000000001, "end": 13.200000000000001, "probability": 1.0 }, { "word": "な", "start": 13.200000000000001, "end": 13.38, "probability": 0.99853515625 }, { "word": "ど", "start": 13.38, "end": 13.58, "probability": 1.0 }, { "word": "電", "start": 13.58, "end": 13.88, "probability": 0.69970703125 }, { "word": "波", "start": 13.88, "end": 14.08, "probability": 0.9970703125 }, { "word": "の", "start": 14.08, "end": 14.22, "probability": 0.99951171875 }, { "word": "出", "start": 14.22, "end": 14.4, "probability": 0.99951171875 }, { "word": "ない", "start": 14.4, "end": 14.620000000000001, "probability": 1.0 }, { "word": "設", "start": 14.620000000000001, "end": 14.9, "probability": 0.99853515625 }, { "word": "定", "start": 14.9, "end": 15.14, "probability": 1.0 }, { "word": "に", "start": 15.14, "end": 15.3, "probability": 1.0 }, { "word": "切", "start": 15.3, "end": 15.46, "probability": 0.99853515625 }, { "word": "り", "start": 15.46, "end": 15.600000000000001, "probability": 0.9990234375 }, { "word": "替", "start": 15.600000000000001, "end": 15.780000000000001, "probability": 0.99951171875 }, { "word": "える", "start": 15.780000000000001, "end": 15.96, "probability": 1.0 }, { "word": "か、", "start": 15.96, "end": 16.18, "probability": 0.99951171875 }, { "word": "電", "start": 16.28, "end": 16.44, "probability": 0.99169921875 }, { "word": "源", "start": 16.44, "end": 16.7, "probability": 0.9990234375 }, { "word": "を", "start": 16.7, "end": 16.919999999999998, "probability": 0.9814453125 }, { "word": "お", "start": 16.919999999999998, "end": 17.1, "probability": 0.99560546875 }, { "word": "切", "start": 17.1, "end": 17.18, "probability": 0.994140625 }, { "word": "り", "start": 17.18, "end": 17.3, "probability": 0.99951171875 }, { "word": "ください。", "start": 17.3, "end": 17.68, "probability": 0.9951171875 } ], "asr_metrics": { "avg_logprob": -0.17755292963098596, "compression_ratio": 1.7377049180327868, "no_speech_prob": 7.893489681332078e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "キャビンアテンダンスドアアウトゥー・ディパーチャーキャビンアテンダンスドアアウトゥー・ディパーチャー皆様、飛行機のドアが閉まりました。電子機器は機内モードなど電波の出ない設定に切り替えるか、電源をお切りください。" } ], "id": 3, "review_annotation": "", "row_id": 3, "bounded_row_id": 3, "parent_row_ids": [ 3 ], "parent_row_id": 3, "packet_id": 0, "LID_evidence": [ { "region_id": 1, "probabilities": { "ja": 0.9779892414808273, "zh": 0.0009155690495390445, "en": 0.012302618621712705 }, "status": "STABLE" }, { "region_id": 2, "probabilities": { "ja": 0.9929139018058777, "zh": 0.00048158445436759695, "en": 0.0029911483458514945 }, "status": "STABLE" } ], "VAD_region_ids": [ 1, 2 ], "language_confidence": 0.9929139018058777, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1.3, "packet_end": 19.0, "raw_text": "電子機器は機内モードなど電波の出ない設定に切り替えるか、電源をお切りください。", "selected_text": "電子機器は機内モードなど電波の出ない設定に切り替えるか、電源をお切りください。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 3 ], "packet_id": 0, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "電子機器は機内モードなど電波の出ない設定に切り替えるか、電源をお切りください。", "after": "電子機器は機内モードなど電波の出ない設定に切り替えるか、電源をお切りください。", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard Japanese announcement text, clearly supported by context and English parallel." } ], "language": "ja", "packet_ids": [ 0 ], "index": 3, "source_text": "電子機器は機内モードなど電波の出ない設定に切り替えるか、電源をお切りください。", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard Japanese announcement text, clearly supported by context and English parallel." }, { "start": 19.3, "end": 21.72, "text": "Ladies and gentlemen, the doors have closed now.", "raw_asr_text": "Ladies and gentlemen, the doors have closed now.", "source_language": "en", "source_lang": "en", "asr_packet_id": 1, "words": [], "asr_metrics": { "avg_logprob": -0.24491043579884064, "compression_ratio": 1.296, "no_speech_prob": 5.481219644831725e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "en", "text": "Ladies and gentlemen, the doors have closed now. Please switch off electronic devices or set them to airplane mode so that radio waves are not emitted. Thank you." } ], "id": 4, "review_annotation": "", "row_id": 4, "bounded_row_id": 4, "parent_row_ids": [ 4 ], "parent_row_id": 4, "packet_id": 1, "LID_evidence": [ { "region_id": 3, "probabilities": { "ja": 0.027452569542219862, "zh": 0.00407997513442145, "en": 0.9267532030741373 }, "status": "STABLE" } ], "VAD_region_ids": [ 3 ], "language_confidence": 0.9267532030741373, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 18.5, "packet_end": 30.7, "raw_text": "Ladies and gentlemen, the doors have closed now.", "selected_text": "Ladies and gentlemen, the doors have closed now.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 4 ], "packet_id": 1, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "Ladies and gentlemen, the doors have closed now.", "after": "Ladies and gentlemen, the doors have closed now.", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard English announcement text, clearly supported by context and Japanese parallel." } ], "language": "en", "packet_ids": [ 1 ], "index": 4, "source_text": "Ladies and gentlemen, the doors have closed now.", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.99, "correction_reason": "Standard English announcement text, clearly supported by context and Japanese parallel.", "raw_word_timestamps": [ { "word": " Ladies", "start": 18.5, "end": 19.64, "probability": 0.6064453125 }, { "word": " and", "start": 19.64, "end": 19.92, "probability": 0.98876953125 }, { "word": " gentlemen,", "start": 19.92, "end": 20.26, "probability": 0.82958984375 }, { "word": " the", "start": 20.58, "end": 20.58, "probability": 0.9873046875 }, { "word": " doors", "start": 20.58, "end": 20.82, "probability": 0.96826171875 }, { "word": " have", "start": 20.82, "end": 21.06, "probability": 0.54541015625 }, { "word": " closed", "start": 21.06, "end": 21.44, "probability": 0.97216796875 }, { "word": " now.", "start": 21.44, "end": 21.72, "probability": 0.9931640625 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 21.9, "end": 28.64, "text": "Please switch off electronic devices or set them to airplane mode so that radio waves are not emitted.", "raw_asr_text": "Please switch off electronic devices or set them to airplane mode so that radio waves are not emitted.", "source_language": "en", "source_lang": "en", "asr_packet_id": 1, "words": [ { "word": " Please", "start": 21.9, "end": 22.240000000000002, "probability": 0.99560546875 }, { "word": " switch", "start": 22.240000000000002, "end": 22.6, "probability": 0.994140625 }, { "word": " off", "start": 22.6, "end": 22.78, "probability": 0.9931640625 }, { "word": " electronic", "start": 22.78, "end": 23.2, "probability": 0.9091796875 }, { "word": " devices", "start": 23.2, "end": 23.86, "probability": 0.99853515625 }, { "word": " or", "start": 23.86, "end": 24.32, "probability": 0.97900390625 }, { "word": " set", "start": 24.32, "end": 24.6, "probability": 0.9931640625 }, { "word": " them", "start": 24.6, "end": 24.8, "probability": 0.9990234375 }, { "word": " to", "start": 24.8, "end": 25.02, "probability": 0.99755859375 }, { "word": " airplane", "start": 25.02, "end": 25.5, "probability": 0.5341796875 }, { "word": " mode", "start": 25.5, "end": 26.0, "probability": 0.9970703125 }, { "word": " so", "start": 26.0, "end": 26.619999999999997, "probability": 0.7705078125 }, { "word": " that", "start": 26.619999999999997, "end": 26.82, "probability": 0.99853515625 }, { "word": " radio", "start": 26.82, "end": 27.22, "probability": 0.98876953125 }, { "word": " waves", "start": 27.22, "end": 27.68, "probability": 0.99951171875 }, { "word": " are", "start": 27.68, "end": 27.9, "probability": 0.9990234375 }, { "word": " not", "start": 27.9, "end": 28.2, "probability": 0.99853515625 }, { "word": " emitted.", "start": 28.2, "end": 28.64, "probability": 0.9970703125 } ], "asr_metrics": { "avg_logprob": -0.24491043579884064, "compression_ratio": 1.296, "no_speech_prob": 5.481219644831725e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "en", "text": "Ladies and gentlemen, the doors have closed now. Please switch off electronic devices or set them to airplane mode so that radio waves are not emitted. Thank you." } ], "id": 5, "review_annotation": "", "row_id": 5, "bounded_row_id": 5, "parent_row_ids": [ 5 ], "parent_row_id": 5, "packet_id": 1, "LID_evidence": [ { "region_id": 3, "probabilities": { "ja": 0.027452569542219862, "zh": 0.00407997513442145, "en": 0.9267532030741373 }, "status": "STABLE" } ], "VAD_region_ids": [ 3 ], "language_confidence": 0.9267532030741373, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 18.5, "packet_end": 30.7, "raw_text": "Please switch off electronic devices or set them to airplane mode so that radio waves are not emitted.", "selected_text": "Please switch off electronic devices or set them to airplane mode so that radio waves are not emitted.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 5 ], "packet_id": 1, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "Please switch off electronic devices or set them to airplane mode so that radio waves are not emitted.", "after": "Please switch off electronic devices or set them to airplane mode so that radio waves are not emitted.", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard English announcement text, clearly supported by context and Japanese parallel." } ], "language": "en", "packet_ids": [ 1 ], "index": 5, "source_text": "Please switch off electronic devices or set them to airplane mode so that radio waves are not emitted.", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard English announcement text, clearly supported by context and Japanese parallel." }, { "start": 28.98, "end": 29.48, "text": "Thank you.", "raw_asr_text": "Thank you.", "source_language": "en", "source_lang": "en", "asr_packet_id": 1, "words": [ { "word": " Thank", "start": 28.98, "end": 29.2, "probability": 0.9951171875 }, { "word": " you.", "start": 29.2, "end": 29.48, "probability": 0.98583984375 } ], "asr_metrics": { "avg_logprob": -0.24491043579884064, "compression_ratio": 1.296, "no_speech_prob": 5.481219644831725e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "en", "text": "Ladies and gentlemen, the doors have closed now. Please switch off electronic devices or set them to airplane mode so that radio waves are not emitted. Thank you." } ], "id": 6, "review_annotation": "", "row_id": 6, "bounded_row_id": 6, "parent_row_ids": [ 6 ], "parent_row_id": 6, "packet_id": 1, "LID_evidence": [ { "region_id": 3, "probabilities": { "ja": 0.027452569542219862, "zh": 0.00407997513442145, "en": 0.9267532030741373 }, "status": "STABLE" } ], "VAD_region_ids": [ 3 ], "language_confidence": 0.9267532030741373, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 18.5, "packet_end": 30.7, "raw_text": "Thank you.", "selected_text": "Thank you.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 6 ], "packet_id": 1, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "Thank you.", "after": "Thank you.", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard English closing, clearly supported by context." } ], "language": "en", "packet_ids": [ 1 ], "index": 6, "source_text": "Thank you.", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.99, "correction_reason": "Standard English closing, clearly supported by context." }, { "start": 32.4, "end": 39.0, "text": "各位貴賓，現在飛機機艙門已經關閉，請您將所有電子電池用品關閉或切換為飛航模式，謝謝", "raw_asr_text": "各位貴賓,現在飛機機艙門已經關閉,請您將所有電子電池用品關閉或切換為飛航模式,謝謝", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 2, "words": [], "asr_metrics": { "avg_logprob": -0.18312161901722784, "compression_ratio": 0.9831932773109243, "no_speech_prob": 1.3953305524894155e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "各位貴賓,現在飛機機艙門已經關閉,請您將所有電子電池用品關閉或切換為飛航模式,謝謝" } ], "id": 7, "review_annotation": "", "row_id": 7, "bounded_row_id": 7, "parent_row_ids": [ 7 ], "parent_row_id": 7, "packet_id": 2, "LID_evidence": [ { "region_id": 4, "probabilities": { "ja": 0.0010936637594237584, "zh": 0.992717520757155, "en": 0.004457358577796681 }, "status": "STABLE" } ], "VAD_region_ids": [ 4 ], "language_confidence": 0.992717520757155, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 31.599999999999998, "packet_end": 40.4, "raw_text": "各位貴賓,現在飛機機艙門已經關閉,請您將所有電子電池用品關閉或切換為飛航模式,謝謝", "selected_text": "各位貴賓，現在飛機機艙門已經關閉，請您將所有電子電池用品關閉或切換為飛航模式，謝謝", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 7 ], "packet_id": 2, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "各位貴賓,現在飛機機艙門已經關閉,請您將所有電子電池用品關閉或切換為飛航模式,謝謝", "after": "各位貴賓，現在飛機機艙門已經關閉，請您將所有電子電池用品關閉或切換為飛航模式，謝謝", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Source text is coherent; '電子電池用品' is likely a specific ASR error for '電子製品' but preserved as per instructions." } ], "language": "zh", "packet_ids": [ 2 ], "index": 7, "independent_ASR": { "text": "各位贵宾，现在飞机机舱门已经关闭，请您将所有电子电器用品关闭或切换为飞行模式，谢谢。", "chunks": [ { "text": "各位贵宾，现在飞机机舱门已经关闭，请您将所有电子电器用品关闭或切换为飞行模式，谢谢。", "start": 0.0, "end": 8.8, "chunk_index": 0, "language": "Chinese", "finish_reason": "eos", "truncated": false, "generated_tokens": 26, "max_new_tokens": 138 } ], "detected_language": "Chinese", "segments": [ { "start": 31.599999999999998, "end": 40.4, "text": "各位贵宾，现在飞机机舱门已经关闭，请您将所有电子电器用品关闭或切换为飞行模式，谢谢。" } ], "start": 31.599999999999998, "end": 40.4, "engine": "mlx-qwen3-asr", "language": "zh" }, "source_text": "各位貴賓，現在飛機機艙門已經關閉，請您將所有電子電池用品關閉或切換為飛航模式，謝謝", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.9, "correction_reason": "Source text is coherent; '電子電池用品' is likely a specific ASR error for '電子製品' but preserved as per instructions.", "raw_word_timestamps": [ { "word": "各位", "start": 32.32, "end": 32.68, "probability": 0.9580078125 }, { "word": "貴", "start": 32.68, "end": 32.879999999999995, "probability": 0.8056640625 }, { "word": "賓,", "start": 32.879999999999995, "end": 33.019999999999996, "probability": 0.998046875 }, { "word": "現在", "start": 33.14, "end": 33.3, "probability": 0.9765625 }, { "word": "飛", "start": 33.3, "end": 33.5, "probability": 0.97265625 }, { "word": "機", "start": 33.5, "end": 33.68, "probability": 0.99755859375 }, { "word": "機", "start": 33.68, "end": 33.8, "probability": 0.99462890625 }, { "word": "艙", "start": 33.8, "end": 34.0, "probability": 0.9443359375 }, { "word": "門", "start": 34.0, "end": 34.18, "probability": 0.67822265625 }, { "word": "已經", "start": 34.18, "end": 34.46, "probability": 0.98291015625 }, { "word": "關", "start": 34.46, "end": 34.68, "probability": 0.9833984375 }, { "word": "閉,", "start": 34.68, "end": 34.96, "probability": 0.9970703125 }, { "word": "請", "start": 34.98, "end": 35.12, "probability": 0.984375 }, { "word": "您", "start": 35.12, "end": 35.239999999999995, "probability": 0.90869140625 }, { "word": "將", "start": 35.239999999999995, "end": 35.419999999999995, "probability": 0.97509765625 }, { "word": "所有", "start": 35.419999999999995, "end": 35.64, "probability": 0.994140625 }, { "word": "電", "start": 35.64, "end": 35.82, "probability": 0.99365234375 }, { "word": "子", "start": 35.82, "end": 35.98, "probability": 0.9990234375 }, { "word": "電", "start": 35.98, "end": 36.12, "probability": 0.998046875 }, { "word": "池", "start": 36.12, "end": 36.32, "probability": 0.78515625 }, { "word": "用", "start": 36.32, "end": 36.48, "probability": 0.98974609375 }, { "word": "品", "start": 36.48, "end": 36.72, "probability": 0.9990234375 }, { "word": "關", "start": 36.72, "end": 36.96, "probability": 0.98095703125 }, { "word": "閉", "start": 36.96, "end": 37.199999999999996, "probability": 0.998046875 }, { "word": "或", "start": 37.199999999999996, "end": 37.36, "probability": 0.98681640625 }, { "word": "切", "start": 37.36, "end": 37.599999999999994, "probability": 0.99755859375 }, { "word": "換", "start": 37.599999999999994, "end": 37.78, "probability": 0.99951171875 }, { "word": "為", "start": 37.78, "end": 37.92, "probability": 0.95654296875 }, { "word": "飛", "start": 37.92, "end": 38.08, "probability": 0.9892578125 }, { "word": "航", "start": 38.08, "end": 38.28, "probability": 0.9521484375 }, { "word": "模", "start": 38.28, "end": 38.44, "probability": 0.9990234375 }, { "word": "式,", "start": 38.44, "end": 38.68, "probability": 1.0 }, { "word": "謝謝", "start": 38.78, "end": 39.0, "probability": 0.9921875 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 42.2, "end": 46.040000000000006, "text": "皆様、スターライアンスメンバーANA東京羽田行き、", "raw_asr_text": "皆様、スターライアンスメンバーANA東京羽田行き、", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 3, "words": [], "asr_metrics": { "avg_logprob": -0.19894316888624622, "compression_ratio": 1.25, "no_speech_prob": 6.125606098750236e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "皆様、スターライアンスメンバーANA東京羽田行き、852便をご利用いただきましてありがとうございます。機長は松井、チーフパーサーは佐藤でございます。" } ], "id": 8, "review_annotation": "", "row_id": 8, "bounded_row_id": 8, "parent_row_ids": [ 8 ], "parent_row_id": 8, "packet_id": 3, "LID_evidence": [ { "region_id": 5, "probabilities": { "ja": 0.9846000763086172, "zh": 0.0011986109143785702, "en": 0.006093612623562177 }, "status": "STABLE" } ], "VAD_region_ids": [ 5 ], "language_confidence": 0.9846000763086172, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 41.400000000000006, "packet_end": 56.9, "raw_text": "皆様、スターライアンスメンバーANA東京羽田行き、", "selected_text": "皆様、スターライアンスメンバーANA東京羽田行き、", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 8 ], "packet_id": 3, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "皆様、スターライアンスメンバーANA東京羽田行き、", "after": "皆様、スターライアンスメンバーANA東京羽田行き、", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard Japanese announcement text, clearly supported by context and English parallel." } ], "language": "ja", "packet_ids": [ 3 ], "index": 8, "source_text": "皆様、スターライアンスメンバーANA東京羽田行き、", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Standard Japanese announcement text, clearly supported by context and English parallel.", "raw_word_timestamps": [ { "word": "皆", "start": 42.06, "end": 42.46000000000001, "probability": 0.9775390625 }, { "word": "様、", "start": 42.46000000000001, "end": 42.720000000000006, "probability": 0.662109375 }, { "word": "ス", "start": 43.00000000000001, "end": 43.06, "probability": 0.9306640625 }, { "word": "ター", "start": 43.06, "end": 43.160000000000004, "probability": 0.98974609375 }, { "word": "ライ", "start": 43.160000000000004, "end": 43.38, "probability": 0.97412109375 }, { "word": "ア", "start": 43.38, "end": 43.50000000000001, "probability": 0.9970703125 }, { "word": "ン", "start": 43.50000000000001, "end": 43.60000000000001, "probability": 0.94482421875 }, { "word": "ス", "start": 43.60000000000001, "end": 43.64000000000001, "probability": 0.99560546875 }, { "word": "メ", "start": 43.64000000000001, "end": 43.82000000000001, "probability": 0.9990234375 }, { "word": "ンバ", "start": 43.82000000000001, "end": 44.00000000000001, "probability": 1.0 }, { "word": "ー", "start": 44.00000000000001, "end": 44.160000000000004, "probability": 1.0 }, { "word": "ANA", "start": 44.160000000000004, "end": 44.580000000000005, "probability": 0.4072265625 }, { "word": "東", "start": 44.580000000000005, "end": 44.980000000000004, "probability": 0.720703125 }, { "word": "京", "start": 44.980000000000004, "end": 45.220000000000006, "probability": 1.0 }, { "word": "羽", "start": 45.220000000000006, "end": 45.46000000000001, "probability": 0.92578125 }, { "word": "田", "start": 45.46000000000001, "end": 45.620000000000005, "probability": 0.99853515625 }, { "word": "行", "start": 45.620000000000005, "end": 45.760000000000005, "probability": 0.74462890625 }, { "word": "き、", "start": 45.760000000000005, "end": 46.040000000000006, "probability": 0.4833984375 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 46.2, "end": 49.540000000000006, "text": "852便をご利用いただきましてありがとうございます。", "raw_asr_text": "852便をご利用いただきましてありがとうございます。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 3, "words": [ { "word": "8", "start": 46.2, "end": 46.50000000000001, "probability": 0.890625 }, { "word": "52", "start": 46.50000000000001, "end": 47.2, "probability": 0.96533203125 }, { "word": "便", "start": 47.2, "end": 47.42, "probability": 0.84619140625 }, { "word": "を", "start": 47.42, "end": 47.660000000000004, "probability": 0.9453125 }, { "word": "ご", "start": 47.660000000000004, "end": 47.84, "probability": 0.99951171875 }, { "word": "利", "start": 47.84, "end": 47.96000000000001, "probability": 1.0 }, { "word": "用", "start": 47.96000000000001, "end": 48.080000000000005, "probability": 1.0 }, { "word": "いただ", "start": 48.080000000000005, "end": 48.28000000000001, "probability": 0.9990234375 }, { "word": "き", "start": 48.28000000000001, "end": 48.52, "probability": 1.0 }, { "word": "ま", "start": 48.52, "end": 48.64000000000001, "probability": 0.99951171875 }, { "word": "して", "start": 48.64000000000001, "end": 48.900000000000006, "probability": 1.0 }, { "word": "ありがとうございます。", "start": 48.900000000000006, "end": 49.540000000000006, "probability": 0.95556640625 } ], "asr_metrics": { "avg_logprob": -0.19894316888624622, "compression_ratio": 1.25, "no_speech_prob": 6.125606098750236e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "皆様、スターライアンスメンバーANA東京羽田行き、852便をご利用いただきましてありがとうございます。機長は松井、チーフパーサーは佐藤でございます。" } ], "id": 9, "review_annotation": "", "row_id": 9, "bounded_row_id": 9, "parent_row_ids": [ 9 ], "parent_row_id": 9, "packet_id": 3, "LID_evidence": [ { "region_id": 5, "probabilities": { "ja": 0.9846000763086172, "zh": 0.0011986109143785702, "en": 0.006093612623562177 }, "status": "STABLE" } ], "VAD_region_ids": [ 5 ], "language_confidence": 0.9846000763086172, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 41.400000000000006, "packet_end": 56.9, "raw_text": "852便をご利用いただきましてありがとうございます。", "selected_text": "852便をご利用いただきましてありがとうございます。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 9 ], "packet_id": 3, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "852便をご利用いただきましてありがとうございます。", "after": "852便をご利用いただきましてありがとうございます。", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard Japanese announcement text, clearly supported by context and English parallel." } ], "language": "ja", "packet_ids": [ 3 ], "index": 9, "source_text": "852便をご利用いただきましてありがとうございます。", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard Japanese announcement text, clearly supported by context and English parallel." }, { "start": 52.080000000000005, "end": 55.760000000000005, "text": "機長は松井、チーフパーサーは佐藤でございます。", "raw_asr_text": "機長は松井、チーフパーサーは佐藤でございます。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 3, "words": [ { "word": "機", "start": 52.080000000000005, "end": 52.160000000000004, "probability": 0.5048828125 }, { "word": "長", "start": 52.160000000000004, "end": 52.36000000000001, "probability": 0.97998046875 }, { "word": "は", "start": 52.36000000000001, "end": 52.64000000000001, "probability": 1.0 }, { "word": "松", "start": 52.64000000000001, "end": 52.980000000000004, "probability": 0.9931640625 }, { "word": "井、", "start": 52.980000000000004, "end": 53.220000000000006, "probability": 0.99853515625 }, { "word": "チ", "start": 53.82000000000001, "end": 53.980000000000004, "probability": 0.95556640625 }, { "word": "ーフ", "start": 53.980000000000004, "end": 54.14000000000001, "probability": 0.9326171875 }, { "word": "パ", "start": 54.14000000000001, "end": 54.28000000000001, "probability": 0.8662109375 }, { "word": "ー", "start": 54.28000000000001, "end": 54.38000000000001, "probability": 0.9677734375 }, { "word": "サ", "start": 54.38000000000001, "end": 54.480000000000004, "probability": 0.986328125 }, { "word": "ー", "start": 54.480000000000004, "end": 54.56, "probability": 0.99853515625 }, { "word": "は", "start": 54.56, "end": 54.74000000000001, "probability": 1.0 }, { "word": "佐", "start": 54.74000000000001, "end": 54.980000000000004, "probability": 0.99853515625 }, { "word": "藤", "start": 54.980000000000004, "end": 55.14000000000001, "probability": 1.0 }, { "word": "で", "start": 55.14000000000001, "end": 55.28000000000001, "probability": 0.99951171875 }, { "word": "ございます。", "start": 55.28000000000001, "end": 55.760000000000005, "probability": 1.0 } ], "asr_metrics": { "avg_logprob": -0.19894316888624622, "compression_ratio": 1.25, "no_speech_prob": 6.125606098750236e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "皆様、スターライアンスメンバーANA東京羽田行き、852便をご利用いただきましてありがとうございます。機長は松井、チーフパーサーは佐藤でございます。" } ], "id": 10, "review_annotation": "待確認", "row_id": 10, "bounded_row_id": 10, "parent_row_ids": [ 10 ], "parent_row_id": 10, "packet_id": 3, "LID_evidence": [ { "region_id": 6, "probabilities": { "ja": 0.8968189358711243, "zh": 0.0035353645325812977, "en": 0.01458782164263539 }, "status": "STABLE" }, { "region_id": 7, "probabilities": { "ja": 0.9126931071281433, "zh": 0.00318483175651636, "en": 0.013760070479474962 }, "status": "STABLE" } ], "VAD_region_ids": [ 6, 7 ], "language_confidence": 0.9126931071281433, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 41.400000000000006, "packet_end": 56.9, "raw_text": "機長は松井、チーフパーサーは佐藤でございます。", "selected_text": "機長は松井、チーフパーサーは佐藤でございます。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 10 ], "packet_id": 3, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "機長は松井、チーフパーサーは佐藤でございます。", "after": "機長は松井、チーフパーサーは佐藤でございます。", "classification": "BEST_SUPPORTED_WITH_REVIEW", "reason": "Independent ASR confirms 'チーフパーサー' (likely 'チーフパーサー' or 'チーフパーサー' misheard for 'チーフパーサー' or 'チーフパーサー')." } ], "language": "ja", "packet_ids": [ 3 ], "index": 10, "independent_ASR": { "text": "皆様、スターライアスメンバーANA東京羽田ゆき852便をご利用いただきましてありがとうございます。機長は松井、チーフパーサーは佐藤でございます。", "chunks": [ { "text": "皆様、スターライアスメンバーANA東京羽田ゆき852便をご利用いただきましてありがとうございます。機長は松井、チーフパーサーは佐藤でございます。", "start": 0.0, "end": 15.5, "chunk_index": 0, "language": "Japanese", "finish_reason": "eos", "truncated": false, "generated_tokens": 46, "max_new_tokens": 218 } ], "detected_language": "Japanese", "segments": [ { "start": 41.400000000000006, "end": 56.900000000000006, "text": "皆様、スターライアスメンバーANA東京羽田ゆき852便をご利用いただきましてありがとうございます。機長は松井、チーフパーサーは佐藤でございます。" } ], "start": 41.400000000000006, "end": 56.9, "engine": "mlx-qwen3-asr", "language": "ja" }, "source_text": "機長は松井、チーフパーサーは佐藤でございます。", "review_status": "BEST_SUPPORTED_WITH_REVIEW", "classification": "BEST_SUPPORTED_WITH_REVIEW", "confidence": 0.8, "correction_reason": "Independent ASR confirms 'チーフパーサー' (likely 'チーフパーサー' or 'チーフパーサー' misheard for 'チーフパーサー' or 'チーフパーサー')." }, { "start": 58.4, "end": 62.54, "text": "ご用の際には遠慮なく客室乗務員にお知らせください。", "raw_asr_text": "雇用の際には遠慮なく客室乗務員にお知らせください。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 4, "words": [], "asr_metrics": { "avg_logprob": -0.1305559299610279, "compression_ratio": 0.9375, "no_speech_prob": 4.789358579865288e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "雇用の際には遠慮なく客室乗務員にお知らせください。" } ], "id": 11, "review_annotation": "待確認", "row_id": 11, "bounded_row_id": 11, "parent_row_ids": [ 11 ], "parent_row_id": 11, "packet_id": 4, "LID_evidence": [ { "region_id": 8, "probabilities": { "ja": 0.991751362880071, "zh": 0.0005858107397216372, "en": 0.003479978690544764 }, "status": "STABLE" } ], "VAD_region_ids": [ 8 ], "language_confidence": 0.991751362880071, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 57.6, "packet_end": 63.9, "raw_text": "雇用の際には遠慮なく客室乗務員にお知らせください。", "selected_text": "ご用の際には遠慮なく客室乗務員にお知らせください。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 11 ], "packet_id": 4, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "雇用の際には遠慮なく客室乗務員にお知らせください。", "after": "ご用の際には遠慮なく客室乗務員にお知らせください。", "classification": "BEST_SUPPORTED_WITH_REVIEW", "reason": "Independent ASR 'お湯' is phonetically 'go-yu', but contextually 'ご用' (go-yō) is the standard phrase for 'needs'." } ], "language": "ja", "packet_ids": [ 4 ], "index": 11, "independent_ASR": { "text": "お湯の際には遠慮なく客室乗務員にお知らせください。", "chunks": [ { "text": "お湯の際には遠慮なく客室乗務員にお知らせください。", "start": 0.0, "end": 6.3, "chunk_index": 0, "language": "Japanese", "finish_reason": "eos", "truncated": false, "generated_tokens": 18, "max_new_tokens": 128 } ], "detected_language": "Japanese", "segments": [ { "start": 57.6, "end": 63.9, "text": "お湯の際には遠慮なく客室乗務員にお知らせください。" } ], "start": 57.6, "end": 63.9, "engine": "mlx-qwen3-asr", "language": "ja" }, "raw_word_timestamps": [ { "word": "雇", "start": 58.18, "end": 58.64, "probability": 0.93115234375 }, { "word": "用", "start": 58.64, "end": 58.78, "probability": 1.0 }, { "word": "の", "start": 58.78, "end": 58.92, "probability": 0.99951171875 }, { "word": "際", "start": 58.92, "end": 59.14, "probability": 1.0 }, { "word": "には", "start": 59.14, "end": 59.480000000000004, "probability": 0.99951171875 }, { "word": "遠", "start": 59.480000000000004, "end": 59.82, "probability": 0.57958984375 }, { "word": "慮", "start": 59.82, "end": 59.94, "probability": 0.99853515625 }, { "word": "なく", "start": 59.94, "end": 60.260000000000005, "probability": 1.0 }, { "word": "客", "start": 60.260000000000005, "end": 60.58, "probability": 0.90673828125 }, { "word": "室", "start": 60.58, "end": 60.88, "probability": 0.99951171875 }, { "word": "乗", "start": 60.88, "end": 61.120000000000005, "probability": 0.98974609375 }, { "word": "務", "start": 61.120000000000005, "end": 61.300000000000004, "probability": 1.0 }, { "word": "員", "start": 61.300000000000004, "end": 61.44, "probability": 0.99755859375 }, { "word": "に", "start": 61.44, "end": 61.660000000000004, "probability": 0.99951171875 }, { "word": "お", "start": 61.660000000000004, "end": 61.9, "probability": 0.998046875 }, { "word": "知", "start": 61.9, "end": 62.0, "probability": 0.99951171875 }, { "word": "ら", "start": 62.0, "end": 62.08, "probability": 1.0 }, { "word": "せ", "start": 62.08, "end": 62.24, "probability": 1.0 }, { "word": "ください。", "start": 62.24, "end": 62.54, "probability": 0.9990234375 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "ご用の際には遠慮なく客室乗務員にお知らせください。", "review_status": "BEST_SUPPORTED_WITH_REVIEW", "classification": "BEST_SUPPORTED_WITH_REVIEW", "confidence": 0.9, "correction_reason": "Independent ASR 'お湯' is phonetically 'go-yu', but contextually 'ご用' (go-yō) is the standard phrase for 'needs'." }, { "start": 68.02, "end": 71.88, "text": "この便は、スターライアンスパートナーとのコードシェア便でございます。", "raw_asr_text": "この便は、スターライアンスパートナーとのコードシェア便でございます。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 5, "words": [ { "word": "この", "start": 68.02, "end": 68.3, "probability": 0.99560546875 }, { "word": "便", "start": 68.3, "end": 68.46000000000001, "probability": 0.966796875 }, { "word": "は、", "start": 68.46000000000001, "end": 68.76, "probability": 0.99853515625 }, { "word": "ス", "start": 68.9, "end": 69.02, "probability": 0.9453125 }, { "word": "ター", "start": 69.02, "end": 69.14, "probability": 0.99853515625 }, { "word": "ライ", "start": 69.14, "end": 69.34, "probability": 0.978515625 }, { "word": "ア", "start": 69.34, "end": 69.52, "probability": 0.9931640625 }, { "word": "ン", "start": 69.52, "end": 69.62, "probability": 0.94091796875 }, { "word": "ス", "start": 69.62, "end": 69.7, "probability": 0.9951171875 }, { "word": "パ", "start": 69.7, "end": 69.86, "probability": 0.9873046875 }, { "word": "ート", "start": 69.86, "end": 70.04, "probability": 1.0 }, { "word": "ナ", "start": 70.04, "end": 70.18, "probability": 0.99951171875 }, { "word": "ー", "start": 70.18, "end": 70.28, "probability": 0.99755859375 }, { "word": "と", "start": 70.28, "end": 70.38000000000001, "probability": 0.99951171875 }, { "word": "の", "start": 70.38000000000001, "end": 70.52, "probability": 1.0 }, { "word": "コ", "start": 70.52, "end": 70.72, "probability": 0.85791015625 }, { "word": "ード", "start": 70.72, "end": 70.86, "probability": 0.99853515625 }, { "word": "シ", "start": 70.86, "end": 71.0, "probability": 0.9931640625 }, { "word": "ェ", "start": 71.0, "end": 71.08, "probability": 0.99853515625 }, { "word": "ア", "start": 71.08, "end": 71.12, "probability": 1.0 }, { "word": "便", "start": 71.12, "end": 71.26, "probability": 0.998046875 }, { "word": "で", "start": 71.26, "end": 71.42, "probability": 0.986328125 }, { "word": "ございます。", "start": 71.42, "end": 71.88, "probability": 1.0 } ], "asr_metrics": { "avg_logprob": -0.14849191052573069, "compression_ratio": 1.1724137931034482, "no_speech_prob": 3.908206223923738e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "この便は、スターライアンスパートナーとのコードシェア便でございます。" } ], "id": 12, "review_annotation": "", "row_id": 12, "bounded_row_id": 12, "parent_row_ids": [ 12 ], "parent_row_id": 12, "packet_id": 5, "LID_evidence": [ { "region_id": 9, "probabilities": { "ja": 0.9785918990770975, "zh": 0.0012962267064722255, "en": 0.009049660982176041 }, "status": "STABLE" } ], "VAD_region_ids": [ 9 ], "language_confidence": 0.9785918990770975, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 67.2, "packet_end": 73.1, "raw_text": "この便は、スターライアンスパートナーとのコードシェア便でございます。", "selected_text": "この便は、スターライアンスパートナーとのコードシェア便でございます。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 12 ], "packet_id": 5, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "この便は、スターライアンスパートナーとのコードシェア便でございます。", "after": "この便は、スターライアンスパートナーとのコードシェア便でございます。", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "日本語原文、英語「Star Wars partner」（Star Allianceの誤聴）および中国語「星空聯盟」の文脈から、原文の「スターライアンス」は「スターアライアンス」の誤記と推測されるが、原文表記を尊重し分類はVALIDとする。" } ], "language": "ja", "packet_ids": [ 5 ], "index": 12, "source_text": "この便は、スターライアンスパートナーとのコードシェア便でございます。", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "日本語原文、英語「Star Wars partner」（Star Allianceの誤聴）および中国語「星空聯盟」の文脈から、原文の「スターライアンス」は「スターアライアンス」の誤記と推測されるが、原文表記を尊重し分類はVALIDとする。" }, { "start": 74.7, "end": 78.4, "text": "前方のスクリーンで非常用設備をご案内いたします。", "raw_asr_text": "前方のスクリーンで非常用設備をご案内いたします。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 6, "words": [], "asr_metrics": { "avg_logprob": -0.1647838679226962, "compression_ratio": 0.9, "no_speech_prob": 4.144635155911569e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "前方のスクリーンで非常用設備をご案内いたします。" } ], "id": 13, "review_annotation": "", "row_id": 13, "bounded_row_id": 13, "parent_row_ids": [ 13 ], "parent_row_id": 13, "packet_id": 6, "LID_evidence": [ { "region_id": 10, "probabilities": { "ja": 0.9724703192710876, "zh": 0.0016003641532734037, "en": 0.011009958572685719 }, "status": "STABLE" } ], "VAD_region_ids": [ 10 ], "language_confidence": 0.9724703192710876, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 73.9, "packet_end": 79.5, "raw_text": "前方のスクリーンで非常用設備をご案内いたします。", "selected_text": "前方のスクリーンで非常用設備をご案内いたします。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 13 ], "packet_id": 6, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "前方のスクリーンで非常用設備をご案内いたします。", "after": "前方のスクリーンで非常用設備をご案内いたします。", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "英語「safety features of this aircraft」および中国語「緊急安全逃生」の文脈と一致し、自然な日本語表現である。" } ], "language": "ja", "packet_ids": [ 6 ], "index": 13, "source_text": "前方のスクリーンで非常用設備をご案内いたします。", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "英語「safety features of this aircraft」および中国語「緊急安全逃生」の文脈と一致し、自然な日本語表現である。", "raw_word_timestamps": [ { "word": "前", "start": 74.68, "end": 75.04, "probability": 0.80615234375 }, { "word": "方", "start": 75.04, "end": 75.2, "probability": 0.9990234375 }, { "word": "の", "start": 75.2, "end": 75.36, "probability": 1.0 }, { "word": "ス", "start": 75.36, "end": 75.56, "probability": 0.99951171875 }, { "word": "ク", "start": 75.56, "end": 75.68, "probability": 1.0 }, { "word": "リ", "start": 75.68, "end": 75.80000000000001, "probability": 1.0 }, { "word": "ーン", "start": 75.80000000000001, "end": 75.92, "probability": 1.0 }, { "word": "で", "start": 75.92, "end": 76.14, "probability": 1.0 }, { "word": "非常", "start": 76.14, "end": 76.48, "probability": 0.89990234375 }, { "word": "用", "start": 76.48, "end": 76.72, "probability": 0.99169921875 }, { "word": "設", "start": 76.72, "end": 76.96000000000001, "probability": 0.99560546875 }, { "word": "備", "start": 76.96000000000001, "end": 77.2, "probability": 0.99951171875 }, { "word": "を", "start": 77.2, "end": 77.36, "probability": 1.0 }, { "word": "ご", "start": 77.36, "end": 77.54, "probability": 0.998046875 }, { "word": "案", "start": 77.54, "end": 77.7, "probability": 0.9990234375 }, { "word": "内", "start": 77.7, "end": 77.86, "probability": 0.99951171875 }, { "word": "いた", "start": 77.86, "end": 78.02000000000001, "probability": 0.99658203125 }, { "word": "します。", "start": 78.02000000000001, "end": 78.4, "probability": 1.0 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 81.3, "end": 90.1, "text": "Good afternoon ladies and gentlemen, welcome on board ANA, a member of Star Alliance flight 852 to Tokyo.", "raw_asr_text": "Good afternoon ladies and gentlemen, welcome on board ANA, a member of Star Lions flight 852 to Tokyo.", "source_language": "en", "source_lang": "en", "asr_packet_id": 7, "words": [], "asr_metrics": { "avg_logprob": -0.2581277953253852, "compression_ratio": 1.386138613861386, "no_speech_prob": 6.645511224645517e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "en", "text": "Good afternoon ladies and gentlemen, welcome on board ANA, a member of Star Lions flight 852 to Tokyo. The captain in command of this flight is Masanori Matsui and I'm chief officer Kaori Sato. Please let us know if you need assistance and we hope you'll enjoy the flight with us." }, { "language": "ja", "text": "Good afternoon ladies and gentlemen welcome on board ANA a member of Star Lions flight 852 to TokyoThe captain in command of this flight is Masanori Matsui and I'm chief officer Kaori SatoPlease let us know if you need assistance and we hope you enjoy the flight with us" } ], "id": 14, "review_annotation": "待確認", "row_id": 14, "bounded_row_id": 14, "parent_row_ids": [ 14 ], "parent_row_id": 14, "packet_id": 7, "LID_evidence": [ { "region_id": 11, "probabilities": { "ja": 0.027614188980526075, "zh": 0.0180976658146112, "en": 0.5034805067743247 }, "status": "UNCERTAIN" }, { "region_id": 12, "probabilities": { "ja": 0.1068168285768479, "zh": 0.008614309877157212, "en": 0.6028351306915283 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 11, 12 ], "language_confidence": 0.6028351306915283, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 80.5, "packet_end": 105.89999999999999, "raw_text": "Good afternoon ladies and gentlemen, welcome on board ANA, a member of Star Lions flight 852 to Tokyo.", "selected_text": "Good afternoon ladies and gentlemen, welcome on board ANA, a member of Star Alliance flight 852 to Tokyo.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 14 ], "packet_id": 7, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "Good afternoon ladies and gentlemen, welcome on board ANA, a member of Star Lions flight 852 to Tokyo.", "after": "Good afternoon ladies and gentlemen, welcome on board ANA, a member of Star Alliance flight 852 to Tokyo.", "classification": "BEST_SUPPORTED_WITH_REVIEW", "reason": "Independent ASR confirms 'Star Alliance'; 'Star Lions' is an ASR error." } ], "language": "en", "packet_ids": [ 7 ], "index": 14, "independent_ASR": { "text": "Good afternoon, ladies and gentlemen. Welcome on board ANA, a member of Star Alliance. Flight 852 to Tokyo. The captain and commander of this flight is Masanori Matsui, and I'm Chief First Class Kauli Sato. Please let us know if you need assistance, and we hope you enjoy the flight with us.", "chunks": [ { "text": "Good afternoon, ladies and gentlemen. Welcome on board ANA, a member of Star Alliance. Flight 852 to Tokyo. The captain and commander of this flight is Masanori Matsui, and I'm Chief First Class Kauli Sato. Please let us know if you need assistance, and we hope you enjoy the flight with us.", "start": 0.0, "end": 25.4, "chunk_index": 0, "language": "English", "finish_reason": "eos", "truncated": false, "generated_tokens": 72, "max_new_tokens": 337 } ], "detected_language": "English", "segments": [ { "start": 80.5, "end": 105.9, "text": "Good afternoon, ladies and gentlemen. Welcome on board ANA, a member of Star Alliance. Flight 852 to Tokyo. The captain and commander of this flight is Masanori Matsui, and I'm Chief First Class Kauli Sato. Please let us know if you need assistance, and we hope you enjoy the flight with us." } ], "start": 80.5, "end": 105.89999999999999, "engine": "mlx-qwen3-asr", "language": "en" }, "raw_word_timestamps": [ { "word": " Good", "start": 80.5, "end": 81.56, "probability": 0.77978515625 }, { "word": " afternoon", "start": 81.56, "end": 81.92, "probability": 0.96923828125 }, { "word": " ladies", "start": 81.92, "end": 82.28, "probability": 0.367431640625 }, { "word": " and", "start": 82.28, "end": 82.62, "probability": 0.98876953125 }, { "word": " gentlemen,", "start": 82.62, "end": 83.02, "probability": 0.78515625 }, { "word": " welcome", "start": 83.32, "end": 83.88, "probability": 0.94677734375 }, { "word": " on", "start": 83.88, "end": 84.08, "probability": 0.7392578125 }, { "word": " board", "start": 84.08, "end": 84.34, "probability": 0.994140625 }, { "word": " ANA,", "start": 84.34, "end": 85.0, "probability": 0.89501953125 }, { "word": " a", "start": 85.28, "end": 85.58, "probability": 0.98779296875 }, { "word": " member", "start": 85.58, "end": 85.84, "probability": 0.99609375 }, { "word": " of", "start": 85.84, "end": 86.12, "probability": 0.9970703125 }, { "word": " Star", "start": 86.12, "end": 86.38, "probability": 0.9462890625 }, { "word": " Lions", "start": 86.38, "end": 86.78, "probability": 0.57421875 }, { "word": " flight", "start": 86.78, "end": 87.18, "probability": 0.68310546875 }, { "word": " 852", "start": 87.18, "end": 89.26, "probability": 0.947265625 }, { "word": " to", "start": 89.26, "end": 89.64, "probability": 0.99169921875 }, { "word": " Tokyo.", "start": 89.64, "end": 90.1, "probability": 0.994140625 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "Good afternoon ladies and gentlemen, welcome on board ANA, a member of Star Alliance flight 852 to Tokyo.", "review_status": "BEST_SUPPORTED_WITH_REVIEW", "classification": "BEST_SUPPORTED_WITH_REVIEW", "confidence": 0.95, "correction_reason": "Independent ASR confirms 'Star Alliance'; 'Star Lions' is an ASR error." }, { "start": 91.22, "end": 97.96000000000001, "text": "The captain in command of this flight is Masanori Matsui and I'm chief officer Kaori Sato.", "raw_asr_text": "The captain in command of this flight is Masanori Matsui and I'm chief officer Kaori Sato.", "source_language": "en", "source_lang": "en", "asr_packet_id": 7, "words": [ { "word": " The", "start": 91.22, "end": 91.82, "probability": 0.9912109375 }, { "word": " captain", "start": 91.82, "end": 92.12, "probability": 0.9638671875 }, { "word": " in", "start": 92.12, "end": 92.4, "probability": 0.9228515625 }, { "word": " command", "start": 92.4, "end": 92.6, "probability": 0.9765625 }, { "word": " of", "start": 92.6, "end": 92.84, "probability": 0.998046875 }, { "word": " this", "start": 92.84, "end": 93.03999999999999, "probability": 0.99462890625 }, { "word": " flight", "start": 93.03999999999999, "end": 93.32, "probability": 0.998046875 }, { "word": " is", "start": 93.32, "end": 93.56, "probability": 0.9990234375 }, { "word": " Masanori", "start": 93.56, "end": 95.22, "probability": 0.86962890625 }, { "word": " Matsui", "start": 95.22, "end": 95.74, "probability": 0.97705078125 }, { "word": " and", "start": 95.74, "end": 96.22, "probability": 0.87353515625 }, { "word": " I'm", "start": 96.22, "end": 96.4, "probability": 0.767578125 }, { "word": " chief", "start": 96.4, "end": 96.64, "probability": 0.426513671875 }, { "word": " officer", "start": 96.64, "end": 97.0, "probability": 0.41162109375 }, { "word": " Kaori", "start": 97.0, "end": 97.5, "probability": 0.9755859375 }, { "word": " Sato.", "start": 97.5, "end": 97.96000000000001, "probability": 0.91259765625 } ], "asr_metrics": { "avg_logprob": -0.2581277953253852, "compression_ratio": 1.386138613861386, "no_speech_prob": 6.645511224645517e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "en", "text": "Good afternoon ladies and gentlemen, welcome on board ANA, a member of Star Lions flight 852 to Tokyo. The captain in command of this flight is Masanori Matsui and I'm chief officer Kaori Sato. Please let us know if you need assistance and we hope you'll enjoy the flight with us." }, { "language": "ja", "text": "Good afternoon ladies and gentlemen welcome on board ANA a member of Star Lions flight 852 to TokyoThe captain in command of this flight is Masanori Matsui and I'm chief officer Kaori SatoPlease let us know if you need assistance and we hope you enjoy the flight with us" } ], "id": 15, "review_annotation": "", "row_id": 15, "bounded_row_id": 15, "parent_row_ids": [ 15 ], "parent_row_id": 15, "packet_id": 7, "LID_evidence": [ { "region_id": 13, "probabilities": { "ja": 0.37643359061330556, "zh": 0.0012786125953425653, "en": 0.5953643015294802 }, "status": "STABLE" }, { "region_id": 14, "probabilities": { "ja": 0.46262718711255324, "zh": 0.00474654627113422, "en": 0.42044989287064205 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 13, 14 ], "language_confidence": 0.5953643015294802, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 80.5, "packet_end": 105.89999999999999, "raw_text": "The captain in command of this flight is Masanori Matsui and I'm chief officer Kaori Sato.", "selected_text": "The captain in command of this flight is Masanori Matsui and I'm chief officer Kaori Sato.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 15 ], "packet_id": 7, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "The captain in command of this flight is Masanori Matsui and I'm chief officer Kaori Sato.", "after": "The captain in command of this flight is Masanori Matsui and I'm chief officer Kaori Sato.", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "日本語「機長は松井、チーフパーサーは佐藤」および中国語「機長は馬次伊、座舱長是Sato」の文脈と一致し、固有名詞も正確。" } ], "language": "en", "packet_ids": [ 7 ], "index": 15, "source_text": "The captain in command of this flight is Masanori Matsui and I'm chief officer Kaori Sato.", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "日本語「機長は松井、チーフパーサーは佐藤」および中国語「機長は馬次伊、座舱長是Sato」の文脈と一致し、固有名詞も正確。" }, { "start": 99.62, "end": 104.84, "text": "Please let us know if you need assistance and we hope you'll enjoy the flight with us.", "raw_asr_text": "Please let us know if you need assistance and we hope you'll enjoy the flight with us.", "source_language": "en", "source_lang": "en", "asr_packet_id": 7, "words": [ { "word": " Please", "start": 99.62, "end": 100.22, "probability": 0.99609375 }, { "word": " let", "start": 100.22, "end": 100.46000000000001, "probability": 0.98779296875 }, { "word": " us", "start": 100.46000000000001, "end": 100.7, "probability": 0.99951171875 }, { "word": " know", "start": 100.7, "end": 100.88, "probability": 0.998046875 }, { "word": " if", "start": 100.88, "end": 101.12, "probability": 0.99853515625 }, { "word": " you", "start": 101.12, "end": 101.22, "probability": 0.99951171875 }, { "word": " need", "start": 101.22, "end": 101.42, "probability": 0.998046875 }, { "word": " assistance", "start": 101.42, "end": 102.03999999999999, "probability": 0.9990234375 }, { "word": " and", "start": 102.03999999999999, "end": 102.88, "probability": 0.896484375 }, { "word": " we", "start": 102.88, "end": 103.02, "probability": 0.99853515625 }, { "word": " hope", "start": 103.02, "end": 103.24, "probability": 0.9990234375 }, { "word": " you'll", "start": 103.24, "end": 103.53999999999999, "probability": 0.64990234375 }, { "word": " enjoy", "start": 103.53999999999999, "end": 103.82, "probability": 0.9990234375 }, { "word": " the", "start": 103.82, "end": 104.0, "probability": 0.9990234375 }, { "word": " flight", "start": 104.0, "end": 104.22, "probability": 0.998046875 }, { "word": " with", "start": 104.22, "end": 104.53999999999999, "probability": 0.99755859375 }, { "word": " us.", "start": 104.53999999999999, "end": 104.84, "probability": 1.0 } ], "asr_metrics": { "avg_logprob": -0.2581277953253852, "compression_ratio": 1.386138613861386, "no_speech_prob": 6.645511224645517e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "en", "text": "Good afternoon ladies and gentlemen, welcome on board ANA, a member of Star Lions flight 852 to Tokyo. The captain in command of this flight is Masanori Matsui and I'm chief officer Kaori Sato. Please let us know if you need assistance and we hope you'll enjoy the flight with us." }, { "language": "ja", "text": "Good afternoon ladies and gentlemen welcome on board ANA a member of Star Lions flight 852 to TokyoThe captain in command of this flight is Masanori Matsui and I'm chief officer Kaori SatoPlease let us know if you need assistance and we hope you enjoy the flight with us" } ], "id": 16, "review_annotation": "", "row_id": 16, "bounded_row_id": 16, "parent_row_ids": [ 16 ], "parent_row_id": 16, "packet_id": 7, "LID_evidence": [ { "region_id": 15, "probabilities": { "ja": 0.0843866379436804, "zh": 0.0037764564651297404, "en": 0.8334573745727539 }, "status": "STABLE" }, { "region_id": 16, "probabilities": { "ja": 0.069131781856413, "zh": 0.00567981121421326, "en": 0.8580140948295594 }, "status": "STABLE" } ], "VAD_region_ids": [ 15, 16 ], "language_confidence": 0.8580140948295594, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 80.5, "packet_end": 105.89999999999999, "raw_text": "Please let us know if you need assistance and we hope you'll enjoy the flight with us.", "selected_text": "Please let us know if you need assistance and we hope you'll enjoy the flight with us.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 16 ], "packet_id": 7, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "Please let us know if you need assistance and we hope you'll enjoy the flight with us.", "after": "Please let us know if you need assistance and we hope you'll enjoy the flight with us.", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "日本語「遠慮なく客室乗務員にお知らせください」および中国語「歡迎隨時用中文與我們聯繫」の文脈と一致し、自然な英語表現である。" } ], "language": "en", "packet_ids": [ 7 ], "index": 16, "source_text": "Please let us know if you need assistance and we hope you'll enjoy the flight with us.", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "日本語「遠慮なく客室乗務員にお知らせください」および中国語「歡迎隨時用中文與我們聯繫」の文脈と一致し、自然な英語表現である。" }, { "start": 107.4, "end": 110.46000000000001, "text": "This is a codeshare flight with our Star Alliance partner.", "raw_asr_text": "This is Korsheir Flight with our Star Wars partner.", "source_language": "en", "source_lang": "en", "asr_packet_id": 8, "words": [], "asr_metrics": { "avg_logprob": -0.33750161257657135, "compression_ratio": 1.2677165354330708, "no_speech_prob": 4.4843049412657265e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "en", "text": "This is Korsheir Flight with our Star Wars partner. We'll now explain the safety features of this aircraft. We ask you for a moment of your attention. Thank you." }, { "language": "ja", "text": "This is Kohl's share flight with our Star Wars partner.We will now explain the safety features of this aircraft.We ask you for a moment of your attention.Thank you." } ], "id": 17, "review_annotation": "待確認", "row_id": 17, "bounded_row_id": 17, "parent_row_ids": [ 17 ], "parent_row_id": 17, "packet_id": 8, "LID_evidence": [ { "region_id": 17, "probabilities": { "ja": 0.0031298871035687625, "zh": 0.0011666864447761327, "en": 0.9808933138847351 }, "status": "STABLE" } ], "VAD_region_ids": [ 17 ], "language_confidence": 0.9808933138847351, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 106.60000000000001, "packet_end": 120.7, "raw_text": "This is Korsheir Flight with our Star Wars partner.", "selected_text": "This is a codeshare flight with our Star Alliance partner.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 17 ], "packet_id": 8, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "This is Korsheir Flight with our Star Wars partner.", "after": "This is a codeshare flight with our Star Alliance partner.", "classification": "BEST_SUPPORTED_WITH_REVIEW", "reason": "Independent ASR 'coach air flight' and 'airlines partner' strongly support 'codeshare flight' and 'Star Alliance'." } ], "language": "en", "packet_ids": [ 8 ], "index": 17, "independent_ASR": { "text": "This is coach air flight with our airlines partner. We'll now explain the safety features of this aircraft. We ask you for a moment of your attention. Thank you.", "chunks": [ { "text": "This is coach air flight with our airlines partner. We'll now explain the safety features of this aircraft. We ask you for a moment of your attention. Thank you.", "start": 0.0, "end": 14.1, "chunk_index": 0, "language": "English", "finish_reason": "eos", "truncated": false, "generated_tokens": 34, "max_new_tokens": 202 } ], "detected_language": "English", "segments": [ { "start": 106.60000000000001, "end": 120.7, "text": "This is coach air flight with our airlines partner. We'll now explain the safety features of this aircraft. We ask you for a moment of your attention. Thank you." } ], "start": 106.60000000000001, "end": 120.7, "engine": "mlx-qwen3-asr", "language": "en" }, "raw_word_timestamps": [ { "word": " This", "start": 106.60000000000001, "end": 107.66000000000001, "probability": 0.84326171875 }, { "word": " is", "start": 107.66000000000001, "end": 107.84, "probability": 0.9814453125 }, { "word": " Korsheir", "start": 107.84, "end": 108.42, "probability": 0.357666015625 }, { "word": " Flight", "start": 108.42, "end": 108.68, "probability": 0.66015625 }, { "word": " with", "start": 108.68, "end": 109.22000000000001, "probability": 0.9580078125 }, { "word": " our", "start": 109.22000000000001, "end": 109.44000000000001, "probability": 0.99365234375 }, { "word": " Star", "start": 109.44000000000001, "end": 109.76, "probability": 0.716796875 }, { "word": " Wars", "start": 109.76, "end": 109.98, "probability": 0.171875 }, { "word": " partner.", "start": 109.98, "end": 110.46000000000001, "probability": 0.89013671875 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "This is a codeshare flight with our Star Alliance partner.", "review_status": "BEST_SUPPORTED_WITH_REVIEW", "classification": "BEST_SUPPORTED_WITH_REVIEW", "confidence": 0.9, "correction_reason": "Independent ASR 'coach air flight' and 'airlines partner' strongly support 'codeshare flight' and 'Star Alliance'." }, { "start": 111.92000000000002, "end": 115.08000000000001, "text": "We'll now explain the safety features of this aircraft.", "raw_asr_text": "We'll now explain the safety features of this aircraft.", "source_language": "en", "source_lang": "en", "asr_packet_id": 8, "words": [ { "word": " We'll", "start": 111.92000000000002, "end": 112.56, "probability": 0.7529296875 }, { "word": " now", "start": 112.56, "end": 112.74000000000001, "probability": 0.99658203125 }, { "word": " explain", "start": 112.74000000000001, "end": 113.14000000000001, "probability": 0.998046875 }, { "word": " the", "start": 113.14000000000001, "end": 113.30000000000001, "probability": 0.99169921875 }, { "word": " safety", "start": 113.30000000000001, "end": 113.64000000000001, "probability": 0.998046875 }, { "word": " features", "start": 113.64000000000001, "end": 114.14000000000001, "probability": 0.9970703125 }, { "word": " of", "start": 114.14000000000001, "end": 114.42000000000002, "probability": 0.998046875 }, { "word": " this", "start": 114.42000000000002, "end": 114.62, "probability": 0.9951171875 }, { "word": " aircraft.", "start": 114.62, "end": 115.08000000000001, "probability": 0.9951171875 } ], "asr_metrics": { "avg_logprob": -0.33750161257657135, "compression_ratio": 1.2677165354330708, "no_speech_prob": 4.4843049412657265e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "en", "text": "This is Korsheir Flight with our Star Wars partner. We'll now explain the safety features of this aircraft. We ask you for a moment of your attention. Thank you." }, { "language": "ja", "text": "This is Kohl's share flight with our Star Wars partner.We will now explain the safety features of this aircraft.We ask you for a moment of your attention.Thank you." } ], "id": 18, "review_annotation": "", "row_id": 18, "bounded_row_id": 18, "parent_row_ids": [ 18 ], "parent_row_id": 18, "packet_id": 8, "LID_evidence": [ { "region_id": 18, "probabilities": { "ja": 0.03470994631892869, "zh": 0.017590410791204443, "en": 0.8161243540900094 }, "status": "STABLE" } ], "VAD_region_ids": [ 18 ], "language_confidence": 0.8161243540900094, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 106.60000000000001, "packet_end": 120.7, "raw_text": "We'll now explain the safety features of this aircraft.", "selected_text": "We'll now explain the safety features of this aircraft.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 18 ], "packet_id": 8, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "We'll now explain the safety features of this aircraft.", "after": "We'll now explain the safety features of this aircraft.", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "日本語「非常用設備をご案内」および中国語「緊急安全逃生」の文脈と一致し、標準的な航空機内アナウンス表現である。" } ], "language": "en", "packet_ids": [ 8 ], "index": 18, "source_text": "We'll now explain the safety features of this aircraft.", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "日本語「非常用設備をご案内」および中国語「緊急安全逃生」の文脈と一致し、標準的な航空機内アナウンス表現である。" }, { "start": 115.62, "end": 119.42000000000002, "text": "We ask you for a moment of your attention. Thank you.", "raw_asr_text": "We ask you for a moment of your attention. Thank you.", "source_language": "en", "source_lang": "en", "asr_packet_id": 8, "words": [ { "word": " We", "start": 115.62, "end": 116.26, "probability": 0.99267578125 }, { "word": " ask", "start": 116.26, "end": 116.50000000000001, "probability": 0.9697265625 }, { "word": " you", "start": 116.50000000000001, "end": 116.64000000000001, "probability": 0.9609375 }, { "word": " for", "start": 116.64000000000001, "end": 116.84, "probability": 0.9814453125 }, { "word": " a", "start": 116.84, "end": 116.96000000000001, "probability": 0.98095703125 }, { "word": " moment", "start": 116.96000000000001, "end": 117.28, "probability": 0.9990234375 }, { "word": " of", "start": 117.28, "end": 117.72000000000001, "probability": 0.990234375 }, { "word": " your", "start": 117.72000000000001, "end": 117.84, "probability": 0.95068359375 }, { "word": " attention.", "start": 117.84, "end": 118.28, "probability": 0.9990234375 }, { "word": " Thank", "start": 118.7, "end": 119.14000000000001, "probability": 0.990234375 }, { "word": " you.", "start": 119.14000000000001, "end": 119.42000000000002, "probability": 0.97216796875 } ], "asr_metrics": { "avg_logprob": -0.33750161257657135, "compression_ratio": 1.2677165354330708, "no_speech_prob": 4.4843049412657265e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "en", "text": "This is Korsheir Flight with our Star Wars partner. We'll now explain the safety features of this aircraft. We ask you for a moment of your attention. Thank you." }, { "language": "ja", "text": "This is Kohl's share flight with our Star Wars partner.We will now explain the safety features of this aircraft.We ask you for a moment of your attention.Thank you." } ], "id": 19, "review_annotation": "", "row_id": 19, "bounded_row_id": 19, "parent_row_ids": [ 19 ], "parent_row_id": 19, "packet_id": 8, "LID_evidence": [ { "region_id": 19, "probabilities": { "ja": 0.003164102672599256, "zh": 0.0034546589013189077, "en": 0.9199744760990143 }, "status": "STABLE" }, { "region_id": 20, "probabilities": { "ja": 0.00488634523935616, "zh": 0.005444573471322656, "en": 0.8323828279972076 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 19, 20 ], "language_confidence": 0.9199744760990143, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 106.60000000000001, "packet_end": 120.7, "raw_text": "We ask you for a moment of your attention. Thank you.", "selected_text": "We ask you for a moment of your attention. Thank you.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 19 ], "packet_id": 8, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "We ask you for a moment of your attention. Thank you.", "after": "We ask you for a moment of your attention. Thank you.", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "日本語「ご案内いたします」および中国語「仔細收看」の文脈と一致し、自然な英語表現である。" } ], "language": "en", "packet_ids": [ 8 ], "index": 19, "source_text": "We ask you for a moment of your attention. Thank you.", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "日本語「ご案内いたします」および中国語「仔細收看」の文脈と一致し、自然な英語表現である。" }, { "start": 123.8, "end": 124.86, "text": "各位貴賓午安", "raw_asr_text": "各位貴賓午安", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 9, "words": [], "asr_metrics": { "avg_logprob": -0.22234791885187596, "compression_ratio": 1.0875, "no_speech_prob": 1.3950711245935832e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "各位貴賓午安歡迎搭乘星空聯盟成員ANA飛往東京的NH-852次班機本班機是ANA與船航航公司的代碼公用班機本班機的機長是馬次伊座舱長" } ], "id": 20, "review_annotation": "", "row_id": 20, "bounded_row_id": 20, "parent_row_ids": [ 20 ], "parent_row_id": 20, "packet_id": 9, "LID_evidence": [ { "region_id": 21, "probabilities": { "ja": 0.008173870854079723, "zh": 0.9313755830128988, "en": 0.04420624921719233 }, "status": "STABLE" } ], "VAD_region_ids": [ 21 ], "language_confidence": 0.9313755830128988, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 123.0, "packet_end": 139.9, "raw_text": "各位貴賓午安", "selected_text": "各位貴賓午安", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 20 ], "packet_id": 9, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "各位貴賓午安", "after": "各位貴賓午安", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard greeting matches context and raw ASR perfectly." } ], "language": "zh", "packet_ids": [ 9 ], "index": 20, "source_text": "各位貴賓午安", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Standard greeting matches context and raw ASR perfectly.", "raw_word_timestamps": [ { "word": "各位", "start": 123.6, "end": 124.04, "probability": 0.990234375 }, { "word": "貴", "start": 124.04, "end": 124.22, "probability": 0.69921875 }, { "word": "賓", "start": 124.22, "end": 124.38, "probability": 0.99755859375 }, { "word": "午", "start": 124.38, "end": 124.62, "probability": 0.599609375 }, { "word": "安", "start": 124.62, "end": 124.86, "probability": 1.0 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 124.86, "end": 130.84, "text": "歡迎搭乘星空聯盟成員ANA飛往東京的NH-852次班機", "raw_asr_text": "歡迎搭乘星空聯盟成員ANA飛往東京的NH-852次班機", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 9, "words": [ { "word": "歡迎", "start": 124.86, "end": 125.58, "probability": 0.2353515625 }, { "word": "搭", "start": 125.58, "end": 125.84, "probability": 0.7744140625 }, { "word": "乘", "start": 125.84, "end": 126.12, "probability": 0.9990234375 }, { "word": "星", "start": 126.12, "end": 126.32, "probability": 0.89111328125 }, { "word": "空", "start": 126.32, "end": 126.56, "probability": 0.99560546875 }, { "word": "聯", "start": 126.56, "end": 126.76, "probability": 0.9853515625 }, { "word": "盟", "start": 126.76, "end": 126.94, "probability": 0.9970703125 }, { "word": "成", "start": 126.94, "end": 127.12, "probability": 0.958984375 }, { "word": "員", "start": 127.12, "end": 127.38, "probability": 0.98583984375 }, { "word": "ANA", "start": 127.38, "end": 127.9, "probability": 0.85498046875 }, { "word": "飛", "start": 127.9, "end": 128.24, "probability": 0.98779296875 }, { "word": "往", "start": 128.24, "end": 128.48, "probability": 0.98779296875 }, { "word": "東", "start": 128.48, "end": 128.94, "probability": 0.97265625 }, { "word": "京", "start": 128.94, "end": 129.16, "probability": 0.99658203125 }, { "word": "的", "start": 129.16, "end": 129.32, "probability": 0.9677734375 }, { "word": "N", "start": 129.32, "end": 129.52, "probability": 0.94580078125 }, { "word": "H", "start": 129.52, "end": 129.74, "probability": 0.98779296875 }, { "word": "-", "start": 129.74, "end": 129.8, "probability": 0.72509765625 }, { "word": "85", "start": 129.8, "end": 130.02, "probability": 0.67919921875 }, { "word": "2", "start": 130.02, "end": 130.2, "probability": 0.77197265625 }, { "word": "次", "start": 130.2, "end": 130.4, "probability": 0.9833984375 }, { "word": "班", "start": 130.4, "end": 130.56, "probability": 0.8330078125 }, { "word": "機", "start": 130.56, "end": 130.84, "probability": 0.9814453125 } ], "asr_metrics": { "avg_logprob": -0.22234791885187596, "compression_ratio": 1.0875, "no_speech_prob": 1.3950711245935832e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "各位貴賓午安歡迎搭乘星空聯盟成員ANA飛往東京的NH-852次班機本班機是ANA與船航航公司的代碼公用班機本班機的機長是馬次伊座舱長" } ], "id": 21, "review_annotation": "", "row_id": 21, "bounded_row_id": 21, "parent_row_ids": [ 21 ], "parent_row_id": 21, "packet_id": 9, "LID_evidence": [ { "region_id": 21, "probabilities": { "ja": 0.008173870854079723, "zh": 0.9313755830128988, "en": 0.04420624921719233 }, "status": "STABLE" }, { "region_id": 22, "probabilities": { "ja": 0.007671324730229874, "zh": 0.9332424561182658, "en": 0.04634034271972875 }, "status": "STABLE" } ], "VAD_region_ids": [ 21, 22 ], "language_confidence": 0.9332424561182658, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 123.0, "packet_end": 139.9, "raw_text": "歡迎搭乘星空聯盟成員ANA飛往東京的NH-852次班機", "selected_text": "歡迎搭乘星空聯盟成員ANA飛往東京的NH-852次班機", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 21 ], "packet_id": 9, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "歡迎搭乘星空聯盟成員ANA飛往東京的NH-852次班機", "after": "歡迎搭乘星空聯盟成員ANA飛往東京的NH-852次班機", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Matches standard announcement phrasing and raw ASR." } ], "language": "zh", "packet_ids": [ 9 ], "index": 21, "source_text": "歡迎搭乘星空聯盟成員ANA飛往東京的NH-852次班機", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Matches standard announcement phrasing and raw ASR." }, { "start": 131.52, "end": 135.66, "text": "本班機是ANA與長榮航空公司的代碼共用班機", "raw_asr_text": "本班機是ANA與船航航公司的代碼公用班機", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 9, "words": [], "asr_metrics": { "avg_logprob": -0.22234791885187596, "compression_ratio": 1.0875, "no_speech_prob": 1.3950711245935832e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "各位貴賓午安歡迎搭乘星空聯盟成員ANA飛往東京的NH-852次班機本班機是ANA與船航航公司的代碼公用班機本班機的機長是馬次伊座舱長" } ], "id": 22, "review_annotation": "待確認", "row_id": 22, "bounded_row_id": 22, "parent_row_ids": [ 22 ], "parent_row_id": 22, "packet_id": 9, "LID_evidence": [ { "region_id": 23, "probabilities": { "ja": 0.007435022387653589, "zh": 0.9291075926560622, "en": 0.05045652312513154 }, "status": "STABLE" } ], "VAD_region_ids": [ 23 ], "language_confidence": 0.9291075926560622, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 123.0, "packet_end": 139.9, "raw_text": "本班機是ANA與船航航公司的代碼公用班機", "selected_text": "本班機是ANA與長榮航空公司的代碼共用班機", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 22 ], "packet_id": 9, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "本班機是ANA與船航航公司的代碼公用班機", "after": "本班機是ANA與長榮航空公司的代碼共用班機", "classification": "BEST_SUPPORTED_WITH_REVIEW", "reason": "Independent ASR '長榮航空公司' (EVA Air) supports the correction from '船航' (nonsense)." } ], "language": "zh", "packet_ids": [ 9 ], "index": 22, "independent_ASR": { "text": "各位贵宾午安，欢迎搭乘星空联盟成员ANA飞往东京的NH八五二次班机。本班机是ANA与长荣航空公司的代码共用班机。本班机的机长是马兹，座舱。", "chunks": [ { "text": "各位贵宾午安，欢迎搭乘星空联盟成员ANA飞往东京的NH八五二次班机。本班机是ANA与长荣航空公司的代码共用班机。本班机的机长是马兹，座舱。", "start": 0.0, "end": 16.9, "chunk_index": 0, "language": "Chinese", "finish_reason": "eos", "truncated": false, "generated_tokens": 52, "max_new_tokens": 235 } ], "detected_language": "Chinese", "segments": [ { "start": 123.0, "end": 139.9, "text": "各位贵宾午安，欢迎搭乘星空联盟成员ANA飞往东京的NH八五二次班机。本班机是ANA与长荣航空公司的代码共用班机。本班机的机长是马兹，座舱。" } ], "start": 123.0, "end": 139.9, "engine": "mlx-qwen3-asr", "language": "zh" }, "raw_word_timestamps": [ { "word": "本", "start": 131.52, "end": 131.96, "probability": 0.96728515625 }, { "word": "班", "start": 131.96, "end": 132.1, "probability": 0.98828125 }, { "word": "機", "start": 132.1, "end": 132.26, "probability": 0.9931640625 }, { "word": "是", "start": 132.26, "end": 132.5, "probability": 0.99462890625 }, { "word": "ANA", "start": 132.5, "end": 132.98, "probability": 0.9814453125 }, { "word": "與", "start": 132.98, "end": 133.44, "probability": 0.36767578125 }, { "word": "船", "start": 133.44, "end": 133.66, "probability": 0.7431640625 }, { "word": "航", "start": 133.66, "end": 133.8, "probability": 0.9609375 }, { "word": "航", "start": 133.8, "end": 134.0, "probability": 0.794921875 }, { "word": "公", "start": 134.0, "end": 134.2, "probability": 0.80859375 }, { "word": "司", "start": 134.2, "end": 134.44, "probability": 0.984375 }, { "word": "的", "start": 134.44, "end": 134.56, "probability": 0.9794921875 }, { "word": "代", "start": 134.56, "end": 134.68, "probability": 0.99609375 }, { "word": "碼", "start": 134.68, "end": 134.86, "probability": 0.9990234375 }, { "word": "公", "start": 134.86, "end": 135.0, "probability": 0.73828125 }, { "word": "用", "start": 135.0, "end": 135.18, "probability": 0.99365234375 }, { "word": "班", "start": 135.18, "end": 135.34, "probability": 0.99609375 }, { "word": "機", "start": 135.34, "end": 135.66, "probability": 0.99658203125 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "本班機是ANA與長榮航空公司的代碼共用班機", "review_status": "BEST_SUPPORTED_WITH_REVIEW", "classification": "BEST_SUPPORTED_WITH_REVIEW", "confidence": 0.9, "correction_reason": "Independent ASR '長榮航空公司' (EVA Air) supports the correction from '船航' (nonsense)." }, { "start": 136.58, "end": 138.88, "text": "本班機的機長是松井", "raw_asr_text": "本班機的機長是馬次伊", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 9, "words": [], "asr_metrics": { "avg_logprob": -0.22234791885187596, "compression_ratio": 1.0875, "no_speech_prob": 1.3950711245935832e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "各位貴賓午安歡迎搭乘星空聯盟成員ANA飛往東京的NH-852次班機本班機是ANA與船航航公司的代碼公用班機本班機的機長是馬次伊座舱長" } ], "id": 23, "review_annotation": "待確認", "row_id": 23, "bounded_row_id": 23, "parent_row_ids": [ 23 ], "parent_row_id": 23, "packet_id": 9, "LID_evidence": [ { "region_id": 24, "probabilities": { "ja": 0.006081351268221624, "zh": 0.971967026591301, "en": 0.014878357644192874 }, "status": "STABLE" } ], "VAD_region_ids": [ 24 ], "language_confidence": 0.971967026591301, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 123.0, "packet_end": 139.9, "raw_text": "本班機的機長是馬次伊", "selected_text": "本班機的機長是松井", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 23 ], "packet_id": 9, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "本班機的機長是馬次伊", "after": "本班機的機長是松井", "classification": "BEST_SUPPORTED_WITH_REVIEW", "reason": "Independent ASR '馬茲' is phonetically 'Matsui' (松井). '馬次伊' is an incorrect transcription." } ], "language": "zh", "packet_ids": [ 9 ], "index": 23, "independent_ASR": { "text": "各位贵宾午安，欢迎搭乘星空联盟成员ANA飞往东京的NH八五二次班机。本班机是ANA与长荣航空公司的代码共用班机。本班机的机长是马兹，座舱。", "chunks": [ { "text": "各位贵宾午安，欢迎搭乘星空联盟成员ANA飞往东京的NH八五二次班机。本班机是ANA与长荣航空公司的代码共用班机。本班机的机长是马兹，座舱。", "start": 0.0, "end": 16.9, "chunk_index": 0, "language": "Chinese", "finish_reason": "eos", "truncated": false, "generated_tokens": 52, "max_new_tokens": 235 } ], "detected_language": "Chinese", "segments": [ { "start": 123.0, "end": 139.9, "text": "各位贵宾午安，欢迎搭乘星空联盟成员ANA飞往东京的NH八五二次班机。本班机是ANA与长荣航空公司的代码共用班机。本班机的机长是马兹，座舱。" } ], "start": 123.0, "end": 139.9, "engine": "mlx-qwen3-asr", "language": "zh" }, "raw_word_timestamps": [ { "word": "本", "start": 136.58, "end": 137.02, "probability": 0.99267578125 }, { "word": "班", "start": 137.02, "end": 137.18, "probability": 0.9990234375 }, { "word": "機", "start": 137.18, "end": 137.34, "probability": 0.99658203125 }, { "word": "的", "start": 137.34, "end": 137.48, "probability": 0.9677734375 }, { "word": "機", "start": 137.48, "end": 137.62, "probability": 0.998046875 }, { "word": "長", "start": 137.62, "end": 137.88, "probability": 0.978515625 }, { "word": "是", "start": 137.88, "end": 138.22, "probability": 0.9951171875 }, { "word": "馬", "start": 138.22, "end": 138.44, "probability": 0.74365234375 }, { "word": "次", "start": 138.44, "end": 138.66, "probability": 0.515625 }, { "word": "伊", "start": 138.66, "end": 138.88, "probability": 0.4765625 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "本班機的機長是松井", "review_status": "BEST_SUPPORTED_WITH_REVIEW", "classification": "BEST_SUPPORTED_WITH_REVIEW", "confidence": 0.9, "correction_reason": "Independent ASR '馬茲' is phonetically 'Matsui' (松井). '馬次伊' is an incorrect transcription." }, { "start": 139.3, "end": 140.8, "text": "此一,座舱長是Sato", "raw_asr_text": "此一,座舱長是Sato", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 10, "words": [], "asr_metrics": { "avg_logprob": -0.20547036810235664, "compression_ratio": 1.0809523809523809, "no_speech_prob": 2.4254319039296313e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "此一,座舱長是Sato畢信蓮,本班機上共配有三名台機組員為您服務若您有任何需要,歡迎隨時用中文與我們聯繫現在為您播放機上的緊急安全逃生失知請各位貴賓仔細收看,謝謝" } ], "id": 24, "review_annotation": "待確認", "row_id": 24, "bounded_row_id": 24, "parent_row_ids": [ 24 ], "parent_row_id": 24, "packet_id": 10, "LID_evidence": [ { "region_id": 25, "probabilities": { "ja": 0.0020655538527838265, "zh": 0.9847656110922496, "en": 0.01023569332707363 }, "status": "STABLE" } ], "VAD_region_ids": [ 25 ], "language_confidence": 0.9847656110922496, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 138.5, "packet_end": 155.0, "raw_text": "此一,座舱長是Sato", "selected_text": "此一,座舱長是Sato", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 24 ], "packet_id": 10, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "此一,座舱長是Sato", "after": "此一,座舱長是Sato", "classification": "BEST_SUPPORTED_WITH_REVIEW", "reason": "Independent ASR '沙朵' supports 'Sato'. '此一' is likely 'Chief' or 'Cabin' misheard.; cross-language change rejected without independent audio ASR" } ], "language": "ja", "packet_ids": [ 10 ], "index": 24, "independent_ASR": { "text": "C，座舱长是沙朵毕信莲。本班机上共配有三名台籍组员为您服务。若您有任何需要，欢迎随时用中文与我们联系。现在为您播放机上的紧急安全逃生须知，请各位贵宾仔细收看，谢谢。", "chunks": [ { "text": "C，座舱长是沙朵毕信莲。本班机上共配有三名台籍组员为您服务。若您有任何需要，欢迎随时用中文与我们联系。现在为您播放机上的紧急安全逃生须知，请各位贵宾仔细收看，谢谢。", "start": 0.0, "end": 16.5, "chunk_index": 0, "language": "Chinese", "finish_reason": "eos", "truncated": false, "generated_tokens": 58, "max_new_tokens": 230 } ], "detected_language": "Chinese", "segments": [ { "start": 138.5, "end": 155.0, "text": "C，座舱长是沙朵毕信莲。本班机上共配有三名台籍组员为您服务。若您有任何需要，欢迎随时用中文与我们联系。现在为您播放机上的紧急安全逃生须知，请各位贵宾仔细收看，谢谢。" } ], "start": 138.5, "end": 155.0, "engine": "mlx-qwen3-asr", "language": "zh" }, "source_text": "此一,座舱長是Sato", "review_status": "BEST_SUPPORTED_WITH_REVIEW", "classification": "BEST_SUPPORTED_WITH_REVIEW", "confidence": 0.85, "correction_reason": "Independent ASR '沙朵' supports 'Sato'. '此一' is likely 'Chief' or 'Cabin' misheard.; cross-language change rejected without independent audio ASR", "raw_word_timestamps": [ { "word": "此", "start": 138.5, "end": 138.66, "probability": 0.2432861328125 }, { "word": "一,", "start": 138.66, "end": 138.88, "probability": 0.65966796875 }, { "word": "座", "start": 139.0, "end": 139.56, "probability": 0.9208984375 }, { "word": "舱", "start": 139.56, "end": 139.78, "probability": 0.93359375 }, { "word": "長", "start": 139.78, "end": 140.0, "probability": 0.98681640625 }, { "word": "是", "start": 140.0, "end": 140.34, "probability": 0.72705078125 }, { "word": "S", "start": 140.34, "end": 140.56, "probability": 0.56982421875 }, { "word": "ato", "start": 140.56, "end": 140.8, "probability": 0.9375 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 140.8, "end": 144.76, "text": "畢信蓮,本班機上共配有三名台機組員為您服務", "raw_asr_text": "畢信蓮,本班機上共配有三名台機組員為您服務", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 10, "words": [ { "word": "畢", "start": 140.8, "end": 141.42, "probability": 0.53173828125 }, { "word": "信", "start": 141.42, "end": 141.56, "probability": 0.626953125 }, { "word": "蓮,", "start": 141.56, "end": 141.84, "probability": 0.82470703125 }, { "word": "本", "start": 142.0, "end": 142.3, "probability": 0.99169921875 }, { "word": "班", "start": 142.3, "end": 142.46, "probability": 0.98876953125 }, { "word": "機", "start": 142.46, "end": 142.56, "probability": 0.90478515625 }, { "word": "上", "start": 142.56, "end": 142.74, "probability": 0.962890625 }, { "word": "共", "start": 142.74, "end": 142.9, "probability": 0.4794921875 }, { "word": "配", "start": 142.9, "end": 143.1, "probability": 0.97998046875 }, { "word": "有", "start": 143.1, "end": 143.24, "probability": 0.97705078125 }, { "word": "三", "start": 143.24, "end": 143.4, "probability": 0.87744140625 }, { "word": "名", "start": 143.4, "end": 143.54, "probability": 0.84375 }, { "word": "台", "start": 143.54, "end": 143.66, "probability": 0.9365234375 }, { "word": "機", "start": 143.66, "end": 143.82, "probability": 0.52685546875 }, { "word": "組", "start": 143.82, "end": 143.96, "probability": 0.970703125 }, { "word": "員", "start": 143.96, "end": 144.1, "probability": 0.99609375 }, { "word": "為", "start": 144.1, "end": 144.24, "probability": 0.986328125 }, { "word": "您", "start": 144.24, "end": 144.38, "probability": 0.99365234375 }, { "word": "服", "start": 144.38, "end": 144.56, "probability": 0.998046875 }, { "word": "務", "start": 144.56, "end": 144.76, "probability": 0.99951171875 } ], "asr_metrics": { "avg_logprob": -0.20547036810235664, "compression_ratio": 1.0809523809523809, "no_speech_prob": 2.4254319039296313e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "此一,座舱長是Sato畢信蓮,本班機上共配有三名台機組員為您服務若您有任何需要,歡迎隨時用中文與我們聯繫現在為您播放機上的緊急安全逃生失知請各位貴賓仔細收看,謝謝" } ], "id": 25, "review_annotation": "待確認", "row_id": 25, "bounded_row_id": 25, "parent_row_ids": [ 25 ], "parent_row_id": 25, "packet_id": 10, "LID_evidence": [ { "region_id": 25, "probabilities": { "ja": 0.0020655538527838265, "zh": 0.9847656110922496, "en": 0.01023569332707363 }, "status": "STABLE" } ], "VAD_region_ids": [ 25 ], "language_confidence": 0.9847656110922496, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 138.5, "packet_end": 155.0, "raw_text": "畢信蓮,本班機上共配有三名台機組員為您服務", "selected_text": "畢信蓮,本班機上共配有三名台機組員為您服務", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 25 ], "packet_id": 10, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "畢信蓮,本班機上共配有三名台機組員為您服務", "after": "畢信蓮,本班機上共配有三名台機組員為您服務", "classification": "BEST_SUPPORTED_WITH_REVIEW", "reason": "Independent ASR identifies '座舱长' and '沙藤' (Sato) preceding '畢信蓮'. Context confirms Chief Purser Sato.; cross-language change rejected without independent audio ASR" } ], "language": "ja", "packet_ids": [ 10 ], "index": 25, "independent_ASR": { "text": "C，座舱长是沙朵毕信莲。本班机上共配有三名台籍组员为您服务。若您有任何需要，欢迎随时用中文与我们联系。现在为您播放机上的紧急安全逃生须知，请各位贵宾仔细收看，谢谢。", "chunks": [ { "text": "C，座舱长是沙朵毕信莲。本班机上共配有三名台籍组员为您服务。若您有任何需要，欢迎随时用中文与我们联系。现在为您播放机上的紧急安全逃生须知，请各位贵宾仔细收看，谢谢。", "start": 0.0, "end": 16.5, "chunk_index": 0, "language": "Chinese", "finish_reason": "eos", "truncated": false, "generated_tokens": 58, "max_new_tokens": 230 } ], "detected_language": "Chinese", "segments": [ { "start": 138.5, "end": 155.0, "text": "C，座舱长是沙朵毕信莲。本班机上共配有三名台籍组员为您服务。若您有任何需要，欢迎随时用中文与我们联系。现在为您播放机上的紧急安全逃生须知，请各位贵宾仔细收看，谢谢。" } ], "start": 138.5, "end": 155.0, "engine": "mlx-qwen3-asr", "language": "zh" }, "source_text": "畢信蓮,本班機上共配有三名台機組員為您服務", "review_status": "BEST_SUPPORTED_WITH_REVIEW", "classification": "BEST_SUPPORTED_WITH_REVIEW", "confidence": 0.85, "correction_reason": "Independent ASR identifies '座舱长' and '沙藤' (Sato) preceding '畢信蓮'. Context confirms Chief Purser Sato.; cross-language change rejected without independent audio ASR" }, { "start": 144.76, "end": 148.14, "text": "若您有任何需要,歡迎隨時用中文與我們聯繫", "raw_asr_text": "若您有任何需要,歡迎隨時用中文與我們聯繫", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 10, "words": [ { "word": "若", "start": 144.76, "end": 145.32, "probability": 0.7060546875 }, { "word": "您", "start": 145.32, "end": 145.46, "probability": 0.95556640625 }, { "word": "有", "start": 145.46, "end": 145.6, "probability": 0.99609375 }, { "word": "任", "start": 145.6, "end": 145.72, "probability": 0.9970703125 }, { "word": "何", "start": 145.72, "end": 145.88, "probability": 0.99951171875 }, { "word": "需要,", "start": 145.88, "end": 146.16, "probability": 0.98681640625 }, { "word": "歡迎", "start": 146.24, "end": 146.46, "probability": 0.9931640625 }, { "word": "隨", "start": 146.46, "end": 146.64, "probability": 0.9833984375 }, { "word": "時", "start": 146.64, "end": 146.78, "probability": 0.9638671875 }, { "word": "用", "start": 146.78, "end": 146.9, "probability": 0.9921875 }, { "word": "中", "start": 146.9, "end": 147.08, "probability": 0.99609375 }, { "word": "文", "start": 147.08, "end": 147.28, "probability": 0.9990234375 }, { "word": "與", "start": 147.28, "end": 147.42, "probability": 0.984375 }, { "word": "我們", "start": 147.42, "end": 147.64, "probability": 0.99267578125 }, { "word": "聯", "start": 147.64, "end": 147.82, "probability": 0.98388671875 }, { "word": "繫", "start": 147.82, "end": 148.14, "probability": 0.96826171875 } ], "asr_metrics": { "avg_logprob": -0.20547036810235664, "compression_ratio": 1.0809523809523809, "no_speech_prob": 2.4254319039296313e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "此一,座舱長是Sato畢信蓮,本班機上共配有三名台機組員為您服務若您有任何需要,歡迎隨時用中文與我們聯繫現在為您播放機上的緊急安全逃生失知請各位貴賓仔細收看,謝謝" } ], "id": 26, "review_annotation": "", "row_id": 26, "bounded_row_id": 26, "parent_row_ids": [ 26 ], "parent_row_id": 26, "packet_id": 10, "LID_evidence": [ { "region_id": 25, "probabilities": { "ja": 0.0020655538527838265, "zh": 0.9847656110922496, "en": 0.01023569332707363 }, "status": "STABLE" } ], "VAD_region_ids": [ 25 ], "language_confidence": 0.9847656110922496, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 138.5, "packet_end": 155.0, "raw_text": "若您有任何需要,歡迎隨時用中文與我們聯繫", "selected_text": "若您有任何需要,歡迎隨時用中文與我們聯繫", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 26 ], "packet_id": 10, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "若您有任何需要,歡迎隨時用中文與我們聯繫", "after": "若您有任何需要,歡迎隨時用中文與我們聯繫", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Matches standard announcement phrasing and raw ASR." } ], "language": "zh", "packet_ids": [ 10 ], "index": 26, "source_text": "若您有任何需要,歡迎隨時用中文與我們聯繫", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Matches standard announcement phrasing and raw ASR." }, { "start": 148.92, "end": 151.86, "text": "現在為您播放機上的緊急安全逃生須知", "raw_asr_text": "現在為您播放機上的緊急安全逃生失知", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 10, "words": [], "asr_metrics": { "avg_logprob": -0.20547036810235664, "compression_ratio": 1.0809523809523809, "no_speech_prob": 2.4254319039296313e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "此一,座舱長是Sato畢信蓮,本班機上共配有三名台機組員為您服務若您有任何需要,歡迎隨時用中文與我們聯繫現在為您播放機上的緊急安全逃生失知請各位貴賓仔細收看,謝謝" } ], "id": 27, "review_annotation": "", "row_id": 27, "bounded_row_id": 27, "parent_row_ids": [ 27 ], "parent_row_id": 27, "packet_id": 10, "LID_evidence": [ { "region_id": 25, "probabilities": { "ja": 0.0020655538527838265, "zh": 0.9847656110922496, "en": 0.01023569332707363 }, "status": "STABLE" } ], "VAD_region_ids": [ 25 ], "language_confidence": 0.9847656110922496, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 138.5, "packet_end": 155.0, "raw_text": "現在為您播放機上的緊急安全逃生失知", "selected_text": "現在為您播放機上的緊急安全逃生須知", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 27 ], "packet_id": 10, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "現在為您播放機上的緊急安全逃生失知", "after": "現在為您播放機上的緊急安全逃生須知", "classification": "CERTAIN_RESTORATION", "reason": "Independent ASR confirms '須知' (instructions) correcting the ASR nonsense '失知'." } ], "language": "zh", "packet_ids": [ 10 ], "index": 27, "independent_ASR": { "text": "C，座舱长是沙朵毕信莲。本班机上共配有三名台籍组员为您服务。若您有任何需要，欢迎随时用中文与我们联系。现在为您播放机上的紧急安全逃生须知，请各位贵宾仔细收看，谢谢。", "chunks": [ { "text": "C，座舱长是沙朵毕信莲。本班机上共配有三名台籍组员为您服务。若您有任何需要，欢迎随时用中文与我们联系。现在为您播放机上的紧急安全逃生须知，请各位贵宾仔细收看，谢谢。", "start": 0.0, "end": 16.5, "chunk_index": 0, "language": "Chinese", "finish_reason": "eos", "truncated": false, "generated_tokens": 58, "max_new_tokens": 230 } ], "detected_language": "Chinese", "segments": [ { "start": 138.5, "end": 155.0, "text": "C，座舱长是沙朵毕信莲。本班机上共配有三名台籍组员为您服务。若您有任何需要，欢迎随时用中文与我们联系。现在为您播放机上的紧急安全逃生须知，请各位贵宾仔细收看，谢谢。" } ], "start": 138.5, "end": 155.0, "engine": "mlx-qwen3-asr", "language": "zh" }, "raw_word_timestamps": [ { "word": "現在", "start": 148.92, "end": 149.24, "probability": 0.3603515625 }, { "word": "為", "start": 149.24, "end": 149.44, "probability": 0.99755859375 }, { "word": "您", "start": 149.44, "end": 149.58, "probability": 0.99853515625 }, { "word": "播", "start": 149.58, "end": 149.74, "probability": 0.99755859375 }, { "word": "放", "start": 149.74, "end": 150.04, "probability": 0.99951171875 }, { "word": "機", "start": 150.04, "end": 150.22, "probability": 0.93017578125 }, { "word": "上", "start": 150.22, "end": 150.38, "probability": 0.99658203125 }, { "word": "的", "start": 150.38, "end": 150.5, "probability": 0.9833984375 }, { "word": "緊", "start": 150.5, "end": 150.64, "probability": 0.98974609375 }, { "word": "急", "start": 150.64, "end": 150.8, "probability": 0.998046875 }, { "word": "安全", "start": 150.8, "end": 151.02, "probability": 0.998046875 }, { "word": "逃", "start": 151.02, "end": 151.3, "probability": 0.92578125 }, { "word": "生", "start": 151.3, "end": 151.48, "probability": 0.98974609375 }, { "word": "失", "start": 151.48, "end": 151.66, "probability": 0.544921875 }, { "word": "知", "start": 151.66, "end": 151.86, "probability": 0.3876953125 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "現在為您播放機上的緊急安全逃生須知", "review_status": "CERTAIN_RESTORATION", "classification": "CERTAIN_RESTORATION", "confidence": 0.95, "correction_reason": "Independent ASR confirms '須知' (instructions) correcting the ASR nonsense '失知'." }, { "start": 151.86, "end": 153.52, "text": "請各位貴賓仔細收看,謝謝", "raw_asr_text": "請各位貴賓仔細收看,謝謝", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 10, "words": [ { "word": "請", "start": 151.86, "end": 152.1, "probability": 0.279052734375 }, { "word": "各位", "start": 152.1, "end": 152.34, "probability": 0.9970703125 }, { "word": "貴", "start": 152.34, "end": 152.46, "probability": 0.98779296875 }, { "word": "賓", "start": 152.46, "end": 152.58, "probability": 0.9970703125 }, { "word": "仔", "start": 152.58, "end": 152.74, "probability": 0.98291015625 }, { "word": "細", "start": 152.74, "end": 152.88, "probability": 0.99951171875 }, { "word": "收看,", "start": 152.88, "end": 153.2, "probability": 0.99560546875 }, { "word": "謝謝", "start": 153.36, "end": 153.52, "probability": 0.9921875 } ], "asr_metrics": { "avg_logprob": -0.20547036810235664, "compression_ratio": 1.0809523809523809, "no_speech_prob": 2.4254319039296313e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "此一,座舱長是Sato畢信蓮,本班機上共配有三名台機組員為您服務若您有任何需要,歡迎隨時用中文與我們聯繫現在為您播放機上的緊急安全逃生失知請各位貴賓仔細收看,謝謝" } ], "id": 28, "review_annotation": "", "row_id": 28, "bounded_row_id": 28, "parent_row_ids": [ 28 ], "parent_row_id": 28, "packet_id": 10, "LID_evidence": [ { "region_id": 25, "probabilities": { "ja": 0.0020655538527838265, "zh": 0.9847656110922496, "en": 0.01023569332707363 }, "status": "STABLE" } ], "VAD_region_ids": [ 25 ], "language_confidence": 0.9847656110922496, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 138.5, "packet_end": 155.0, "raw_text": "請各位貴賓仔細收看,謝謝", "selected_text": "請各位貴賓仔細收看,謝謝", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 28 ], "packet_id": 10, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "請各位貴賓仔細收看,謝謝", "after": "請各位貴賓仔細收看,謝謝", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Matches standard announcement phrasing and raw ASR." } ], "language": "zh", "packet_ids": [ 10 ], "index": 28, "source_text": "請各位貴賓仔細收看,謝謝", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Matches standard announcement phrasing and raw ASR." }, { "start": 178.5, "end": 180.2, "text": "[聴不清／待確認]", "raw_asr_text": "", "source_language": "zh", "source_lang": "zh", "vad_region_id": 26, "review_status": "UNRECOVERABLE_MARKED", "asr_packet_id": 11, "correction_reason": "Audio contains unrelated promotional content; no valid speech recovery exists for the announcement.", "confidence": 0.1, "id": 29, "review_annotation": "待確認", "row_id": 29, "bounded_row_id": 29, "parent_row_ids": [ 29 ], "parent_row_id": 29, "packet_id": 11, "LID_evidence": [ { "region_id": 26, "probabilities": { "ja": 0.08285250730114058, "zh": 0.7573004849255085, "en": 0.08775666914880276 }, "status": "STABLE" } ], "VAD_region_ids": [ 26 ], "language_confidence": 0.7573004849255085, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 177.7, "packet_end": 200.9, "raw_text": "", "selected_text": "[聴不清／待確認]", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 29 ], "packet_id": 11, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "[聴不清／待確認]", "after": "[聴不清／待確認]", "classification": "UNRECOVERABLE_MARKED", "reason": "Audio contains unrelated promotional content; no valid speech recovery exists for the announcement." } ], "language": "zh", "packet_ids": [ 11 ], "index": 29, "independent_ASR": { "engine": "whisper-large-v3", "model_path": "/Users/sagawa/AI_WORK/cache/huggingface/hub/models--mlx-community--whisper-large-v3", "language": "zh", "audio_interval": [ 177.7, 200.9 ], "target_interval": [ 178.5, 200.1 ], "text": "请不吝点赞订阅转发打赏支持明镜与点点栏目", "words": [ { "word": "请", "start": 180.26, "end": 180.85999999999999, "probability": 0.343994140625 }, { "word": "不", "start": 180.85999999999999, "end": 180.85999999999999, "probability": 0.0040740966796875 }, { "word": "吝", "start": 180.85999999999999, "end": 183.45999999999998, "probability": 0.98828125 }, { "word": "点", "start": 183.45999999999998, "end": 183.61999999999998, "probability": 0.9990234375 }, { "word": "赞", "start": 183.61999999999998, "end": 183.66, "probability": 0.9990234375 }, { "word": " 订", "start": 183.66, "end": 183.83999999999997, "probability": 0.9990234375 }, { "word": "阅", "start": 183.83999999999997, "end": 184.0, "probability": 0.99951171875 }, { "word": " 转", "start": 184.0, "end": 184.38, "probability": 0.9990234375 }, { "word": "发", "start": 184.38, "end": 184.38, "probability": 0.99658203125 }, { "word": " 打", "start": 184.45999999999998, "end": 184.56, "probability": 0.9990234375 }, { "word": "赏", "start": 184.56, "end": 191.76, "probability": 0.9990234375 }, { "word": "支持", "start": 191.76, "end": 193.17999999999998, "probability": 0.99951171875 }, { "word": "明", "start": 193.17999999999998, "end": 196.98, "probability": 0.99951171875 }, { "word": "镜", "start": 196.98, "end": 197.95999999999998, "probability": 1.0 }, { "word": "与", "start": 197.95999999999998, "end": 198.23999999999998, "probability": 0.99951171875 }, { "word": "点", "start": 198.23999999999998, "end": 198.54, "probability": 0.99951171875 }, { "word": "点", "start": 198.54, "end": 198.83999999999997, "probability": 0.99951171875 }, { "word": "栏", "start": 198.83999999999997, "end": 199.07999999999998, "probability": 0.99951171875 }, { "word": "目", "start": 199.07999999999998, "end": 200.88, "probability": 1.0 } ], "raw_result": { "text": "请不吝点赞 订阅 转发 打赏支持明镜与点点栏目", "segments": [ { "id": 0, "seek": 0, "start": 2.5599999999999987, "end": 23.18, "text": "请不吝点赞 订阅 转发 打赏支持明镜与点点栏目", "tokens": [ 50365, 27908, 1960, 2392, 251, 12579, 5266, 252, 220, 7422, 95, 10034, 227, 220, 17819, 105, 28926, 220, 12467, 5266, 237, 35488, 11100, 12373, 250, 940, 236, 12579, 12579, 8987, 237, 11386, 51864 ], "temperature": 0.0, "avg_logprob": -0.1714823806987089, "compression_ratio": 0.8513513513513513, "no_speech_prob": 0.04063153639435768, "words": [ { "word": "请", "start": 2.5599999999999987, "end": 3.16, "probability": 0.343994140625 }, { "word": "不", "start": 3.16, "end": 3.16, "probability": 0.0040740966796875 }, { "word": "吝", "start": 3.16, "end": 5.76, "probability": 0.98828125 }, { "word": "点", "start": 5.76, "end": 5.92, "probability": 0.9990234375 }, { "word": "赞", "start": 5.92, "end": 5.96, "probability": 0.9990234375 }, { "word": " 订", "start": 5.96, "end": 6.14, "probability": 0.9990234375 }, { "word": "阅", "start": 6.14, "end": 6.3, "probability": 0.99951171875 }, { "word": " 转", "start": 6.3, "end": 6.68, "probability": 0.9990234375 }, { "word": "发", "start": 6.68, "end": 6.68, "probability": 0.99658203125 }, { "word": " 打", "start": 6.76, "end": 6.86, "probability": 0.9990234375 }, { "word": "赏", "start": 6.86, "end": 14.06, "probability": 0.9990234375 }, { "word": "支持", "start": 14.06, "end": 15.48, "probability": 0.99951171875 }, { "word": "明", "start": 15.48, "end": 19.28, "probability": 0.99951171875 }, { "word": "镜", "start": 19.28, "end": 20.26, "probability": 1.0 }, { "word": "与", "start": 20.26, "end": 20.54, "probability": 0.99951171875 }, { "word": "点", "start": 20.54, "end": 20.84, "probability": 0.99951171875 }, { "word": "点", "start": 20.84, "end": 21.14, "probability": 0.99951171875 }, { "word": "栏", "start": 21.14, "end": 21.38, "probability": 0.99951171875 }, { "word": "目", "start": 21.38, "end": 23.18, "probability": 1.0 } ] } ], "language": "zh", "duration": 23.2 }, "status": "COMPLETE", "rejected_alternative": { "text": "ご到着ありがとうございます。Welcome aboard, A N A。これより機内の安全についてご案内します。May we have your attention for the following safety instructions?荷物は座席の上の物入れまたは前の座席の下に収納してください。Please place your baggage in the overhead compartment or under the seat in front of you.", "chunks": [ { "text": "ご到着ありがとうございます。Welcome aboard, A N A。これより機内の安全についてご案内します。May we have your attention for the following safety instructions?荷物は座席の上の物入れまたは前の座席の下に収納してください。Please place your baggage in the overhead compartment or under the seat in front of you.", "start": 0.0, "end": 23.2, "chunk_index": 0, "language": "Chinese", "finish_reason": "eos", "truncated": false, "generated_tokens": 72, "max_new_tokens": 311 } ], "detected_language": "Chinese", "segments": [ { "start": 177.7, "end": 200.89999999999998, "text": "ご到着ありがとうございます。Welcome aboard, A N A。これより機内の安全についてご案内します。May we have your attention for the following safety instructions?荷物は座席の上の物入れまたは前の座席の下に収納してください。Please place your baggage in the overhead compartment or under the seat in front of you." } ], "start": 177.7, "end": 200.9, "engine": "mlx-qwen3-asr", "language": "zh" } }, "source_text": "[聴不清／待確認]", "classification": "UNRECOVERABLE_MARKED" }, { "start": 180.29999999999998, "end": 182.1, "text": "歡迎來到A N A", "raw_asr_text": "歡迎來到A&A", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 11, "words": [], "asr_metrics": { "avg_logprob": -0.4212241617838542, "compression_ratio": 4.777777777777778, "no_speech_prob": 3.221098041072956e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "zh", "text": "謝謝你搭乘的歡迎來到A&A請介紹一下機的安全我們請你注意安全的指示請坐坐上的物入或是坐坐下的坐坐下拜拜" }, { "language": "en", "text": "Thank you for your ride. Welcome aboard A&A. I will tell you about the safety of the car. Here we have your attention for the following safety instructions. Your luggage is on the seat of the seat. Please put your luggage in the overhead compartment or under the seat in front of you." } ], "id": 30, "review_annotation": "待確認", "row_id": 30, "bounded_row_id": 30, "parent_row_ids": [ 30 ], "parent_row_id": 30, "packet_id": 11, "LID_evidence": [ { "region_id": 27, "probabilities": { "ja": 0.015955725335516036, "zh": 0.7250325101028595, "en": 0.10977546185521143 }, "status": "STABLE" } ], "VAD_region_ids": [ 27 ], "language_confidence": 0.7250325101028595, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 177.7, "packet_end": 200.9, "raw_text": "歡迎來到A&A", "selected_text": "歡迎來到A N A", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 30 ], "packet_id": 11, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "歡迎來到A&A", "after": "歡迎來到A N A", "classification": "BEST_SUPPORTED_WITH_REVIEW", "reason": "Context and rejected alternative confirm 'A N A' (ANA) rather than 'A&A'." } ], "language": "zh", "packet_ids": [ 11 ], "index": 30, "independent_ASR": { "engine": "whisper-large-v3", "model_path": "/Users/sagawa/AI_WORK/cache/huggingface/hub/models--mlx-community--whisper-large-v3", "language": "zh", "audio_interval": [ 177.7, 200.9 ], "target_interval": [ 178.5, 200.1 ], "text": "请不吝点赞订阅转发打赏支持明镜与点点栏目", "words": [ { "word": "请", "start": 180.26, "end": 180.85999999999999, "probability": 0.343994140625 }, { "word": "不", "start": 180.85999999999999, "end": 180.85999999999999, "probability": 0.0040740966796875 }, { "word": "吝", "start": 180.85999999999999, "end": 183.45999999999998, "probability": 0.98828125 }, { "word": "点", "start": 183.45999999999998, "end": 183.61999999999998, "probability": 0.9990234375 }, { "word": "赞", "start": 183.61999999999998, "end": 183.66, "probability": 0.9990234375 }, { "word": " 订", "start": 183.66, "end": 183.83999999999997, "probability": 0.9990234375 }, { "word": "阅", "start": 183.83999999999997, "end": 184.0, "probability": 0.99951171875 }, { "word": " 转", "start": 184.0, "end": 184.38, "probability": 0.9990234375 }, { "word": "发", "start": 184.38, "end": 184.38, "probability": 0.99658203125 }, { "word": " 打", "start": 184.45999999999998, "end": 184.56, "probability": 0.9990234375 }, { "word": "赏", "start": 184.56, "end": 191.76, "probability": 0.9990234375 }, { "word": "支持", "start": 191.76, "end": 193.17999999999998, "probability": 0.99951171875 }, { "word": "明", "start": 193.17999999999998, "end": 196.98, "probability": 0.99951171875 }, { "word": "镜", "start": 196.98, "end": 197.95999999999998, "probability": 1.0 }, { "word": "与", "start": 197.95999999999998, "end": 198.23999999999998, "probability": 0.99951171875 }, { "word": "点", "start": 198.23999999999998, "end": 198.54, "probability": 0.99951171875 }, { "word": "点", "start": 198.54, "end": 198.83999999999997, "probability": 0.99951171875 }, { "word": "栏", "start": 198.83999999999997, "end": 199.07999999999998, "probability": 0.99951171875 }, { "word": "目", "start": 199.07999999999998, "end": 200.88, "probability": 1.0 } ], "raw_result": { "text": "请不吝点赞 订阅 转发 打赏支持明镜与点点栏目", "segments": [ { "id": 0, "seek": 0, "start": 2.5599999999999987, "end": 23.18, "text": "请不吝点赞 订阅 转发 打赏支持明镜与点点栏目", "tokens": [ 50365, 27908, 1960, 2392, 251, 12579, 5266, 252, 220, 7422, 95, 10034, 227, 220, 17819, 105, 28926, 220, 12467, 5266, 237, 35488, 11100, 12373, 250, 940, 236, 12579, 12579, 8987, 237, 11386, 51864 ], "temperature": 0.0, "avg_logprob": -0.1714823806987089, "compression_ratio": 0.8513513513513513, "no_speech_prob": 0.04063153639435768, "words": [ { "word": "请", "start": 2.5599999999999987, "end": 3.16, "probability": 0.343994140625 }, { "word": "不", "start": 3.16, "end": 3.16, "probability": 0.0040740966796875 }, { "word": "吝", "start": 3.16, "end": 5.76, "probability": 0.98828125 }, { "word": "点", "start": 5.76, "end": 5.92, "probability": 0.9990234375 }, { "word": "赞", "start": 5.92, "end": 5.96, "probability": 0.9990234375 }, { "word": " 订", "start": 5.96, "end": 6.14, "probability": 0.9990234375 }, { "word": "阅", "start": 6.14, "end": 6.3, "probability": 0.99951171875 }, { "word": " 转", "start": 6.3, "end": 6.68, "probability": 0.9990234375 }, { "word": "发", "start": 6.68, "end": 6.68, "probability": 0.99658203125 }, { "word": " 打", "start": 6.76, "end": 6.86, "probability": 0.9990234375 }, { "word": "赏", "start": 6.86, "end": 14.06, "probability": 0.9990234375 }, { "word": "支持", "start": 14.06, "end": 15.48, "probability": 0.99951171875 }, { "word": "明", "start": 15.48, "end": 19.28, "probability": 0.99951171875 }, { "word": "镜", "start": 19.28, "end": 20.26, "probability": 1.0 }, { "word": "与", "start": 20.26, "end": 20.54, "probability": 0.99951171875 }, { "word": "点", "start": 20.54, "end": 20.84, "probability": 0.99951171875 }, { "word": "点", "start": 20.84, "end": 21.14, "probability": 0.99951171875 }, { "word": "栏", "start": 21.14, "end": 21.38, "probability": 0.99951171875 }, { "word": "目", "start": 21.38, "end": 23.18, "probability": 1.0 } ] } ], "language": "zh", "duration": 23.2 }, "status": "COMPLETE", "rejected_alternative": { "text": "ご到着ありがとうございます。Welcome aboard, A N A。これより機内の安全についてご案内します。May we have your attention for the following safety instructions?荷物は座席の上の物入れまたは前の座席の下に収納してください。Please place your baggage in the overhead compartment or under the seat in front of you.", "chunks": [ { "text": "ご到着ありがとうございます。Welcome aboard, A N A。これより機内の安全についてご案内します。May we have your attention for the following safety instructions?荷物は座席の上の物入れまたは前の座席の下に収納してください。Please place your baggage in the overhead compartment or under the seat in front of you.", "start": 0.0, "end": 23.2, "chunk_index": 0, "language": "Chinese", "finish_reason": "eos", "truncated": false, "generated_tokens": 72, "max_new_tokens": 311 } ], "detected_language": "Chinese", "segments": [ { "start": 177.7, "end": 200.89999999999998, "text": "ご到着ありがとうございます。Welcome aboard, A N A。これより機内の安全についてご案内します。May we have your attention for the following safety instructions?荷物は座席の上の物入れまたは前の座席の下に収納してください。Please place your baggage in the overhead compartment or under the seat in front of you." } ], "start": 177.7, "end": 200.9, "engine": "mlx-qwen3-asr", "language": "zh" } }, "raw_word_timestamps": [ { "word": "歡迎", "start": 180.29999999999998, "end": 180.83999999999997, "probability": 0.0025806427001953125 }, { "word": "來到", "start": 180.83999999999997, "end": 181.38, "probability": 0.1966552734375 }, { "word": "A", "start": 181.38, "end": 181.67999999999998, "probability": 0.1759033203125 }, { "word": "&", "start": 181.67999999999998, "end": 181.82, "probability": 0.53515625 }, { "word": "A", "start": 181.82, "end": 182.1, "probability": 0.99658203125 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "歡迎來到A N A", "review_status": "BEST_SUPPORTED_WITH_REVIEW", "classification": "BEST_SUPPORTED_WITH_REVIEW", "confidence": 0.8, "correction_reason": "Context and rejected alternative confirm 'A N A' (ANA) rather than 'A&A'." }, { "start": 182.88, "end": 184.38, "text": "請介紹機內的安全", "raw_asr_text": "請介紹一下機的安全", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 11, "words": [], "asr_metrics": { "avg_logprob": -0.4212241617838542, "compression_ratio": 4.777777777777778, "no_speech_prob": 3.221098041072956e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "zh", "text": "謝謝你搭乘的歡迎來到A&A請介紹一下機的安全我們請你注意安全的指示請坐坐上的物入或是坐坐下的坐坐下拜拜" }, { "language": "en", "text": "Thank you for your ride. Welcome aboard A&A. I will tell you about the safety of the car. Here we have your attention for the following safety instructions. Your luggage is on the seat of the seat. Please put your luggage in the overhead compartment or under the seat in front of you." } ], "id": 31, "review_annotation": "待確認", "row_id": 31, "bounded_row_id": 31, "parent_row_ids": [ 31 ], "parent_row_id": 31, "packet_id": 11, "LID_evidence": [ { "region_id": 28, "probabilities": { "ja": 0.14047966427824576, "zh": 0.07362690406625916, "en": 0.3789523937230115 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 28 ], "language_confidence": 0.07362690406625916, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 177.7, "packet_end": 200.9, "raw_text": "請介紹一下機的安全", "selected_text": "請介紹機內的安全", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 31 ], "packet_id": 11, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "請介紹一下機的安全", "after": "請介紹機內的安全", "classification": "BEST_SUPPORTED_WITH_REVIEW", "reason": "Context '機内的安全' and rejected alternative support '機內' over '機'." } ], "language": "zh", "packet_ids": [ 11 ], "index": 31, "independent_ASR": { "engine": "whisper-large-v3", "model_path": "/Users/sagawa/AI_WORK/cache/huggingface/hub/models--mlx-community--whisper-large-v3", "language": "zh", "audio_interval": [ 177.7, 200.9 ], "target_interval": [ 178.5, 200.1 ], "text": "请不吝点赞订阅转发打赏支持明镜与点点栏目", "words": [ { "word": "请", "start": 180.26, "end": 180.85999999999999, "probability": 0.343994140625 }, { "word": "不", "start": 180.85999999999999, "end": 180.85999999999999, "probability": 0.0040740966796875 }, { "word": "吝", "start": 180.85999999999999, "end": 183.45999999999998, "probability": 0.98828125 }, { "word": "点", "start": 183.45999999999998, "end": 183.61999999999998, "probability": 0.9990234375 }, { "word": "赞", "start": 183.61999999999998, "end": 183.66, "probability": 0.9990234375 }, { "word": " 订", "start": 183.66, "end": 183.83999999999997, "probability": 0.9990234375 }, { "word": "阅", "start": 183.83999999999997, "end": 184.0, "probability": 0.99951171875 }, { "word": " 转", "start": 184.0, "end": 184.38, "probability": 0.9990234375 }, { "word": "发", "start": 184.38, "end": 184.38, "probability": 0.99658203125 }, { "word": " 打", "start": 184.45999999999998, "end": 184.56, "probability": 0.9990234375 }, { "word": "赏", "start": 184.56, "end": 191.76, "probability": 0.9990234375 }, { "word": "支持", "start": 191.76, "end": 193.17999999999998, "probability": 0.99951171875 }, { "word": "明", "start": 193.17999999999998, "end": 196.98, "probability": 0.99951171875 }, { "word": "镜", "start": 196.98, "end": 197.95999999999998, "probability": 1.0 }, { "word": "与", "start": 197.95999999999998, "end": 198.23999999999998, "probability": 0.99951171875 }, { "word": "点", "start": 198.23999999999998, "end": 198.54, "probability": 0.99951171875 }, { "word": "点", "start": 198.54, "end": 198.83999999999997, "probability": 0.99951171875 }, { "word": "栏", "start": 198.83999999999997, "end": 199.07999999999998, "probability": 0.99951171875 }, { "word": "目", "start": 199.07999999999998, "end": 200.88, "probability": 1.0 } ], "raw_result": { "text": "请不吝点赞 订阅 转发 打赏支持明镜与点点栏目", "segments": [ { "id": 0, "seek": 0, "start": 2.5599999999999987, "end": 23.18, "text": "请不吝点赞 订阅 转发 打赏支持明镜与点点栏目", "tokens": [ 50365, 27908, 1960, 2392, 251, 12579, 5266, 252, 220, 7422, 95, 10034, 227, 220, 17819, 105, 28926, 220, 12467, 5266, 237, 35488, 11100, 12373, 250, 940, 236, 12579, 12579, 8987, 237, 11386, 51864 ], "temperature": 0.0, "avg_logprob": -0.1714823806987089, "compression_ratio": 0.8513513513513513, "no_speech_prob": 0.04063153639435768, "words": [ { "word": "请", "start": 2.5599999999999987, "end": 3.16, "probability": 0.343994140625 }, { "word": "不", "start": 3.16, "end": 3.16, "probability": 0.0040740966796875 }, { "word": "吝", "start": 3.16, "end": 5.76, "probability": 0.98828125 }, { "word": "点", "start": 5.76, "end": 5.92, "probability": 0.9990234375 }, { "word": "赞", "start": 5.92, "end": 5.96, "probability": 0.9990234375 }, { "word": " 订", "start": 5.96, "end": 6.14, "probability": 0.9990234375 }, { "word": "阅", "start": 6.14, "end": 6.3, "probability": 0.99951171875 }, { "word": " 转", "start": 6.3, "end": 6.68, "probability": 0.9990234375 }, { "word": "发", "start": 6.68, "end": 6.68, "probability": 0.99658203125 }, { "word": " 打", "start": 6.76, "end": 6.86, "probability": 0.9990234375 }, { "word": "赏", "start": 6.86, "end": 14.06, "probability": 0.9990234375 }, { "word": "支持", "start": 14.06, "end": 15.48, "probability": 0.99951171875 }, { "word": "明", "start": 15.48, "end": 19.28, "probability": 0.99951171875 }, { "word": "镜", "start": 19.28, "end": 20.26, "probability": 1.0 }, { "word": "与", "start": 20.26, "end": 20.54, "probability": 0.99951171875 }, { "word": "点", "start": 20.54, "end": 20.84, "probability": 0.99951171875 }, { "word": "点", "start": 20.84, "end": 21.14, "probability": 0.99951171875 }, { "word": "栏", "start": 21.14, "end": 21.38, "probability": 0.99951171875 }, { "word": "目", "start": 21.38, "end": 23.18, "probability": 1.0 } ] } ], "language": "zh", "duration": 23.2 }, "status": "COMPLETE", "rejected_alternative": { "text": "ご到着ありがとうございます。Welcome aboard, A N A。これより機内の安全についてご案内します。May we have your attention for the following safety instructions?荷物は座席の上の物入れまたは前の座席の下に収納してください。Please place your baggage in the overhead compartment or under the seat in front of you.", "chunks": [ { "text": "ご到着ありがとうございます。Welcome aboard, A N A。これより機内の安全についてご案内します。May we have your attention for the following safety instructions?荷物は座席の上の物入れまたは前の座席の下に収納してください。Please place your baggage in the overhead compartment or under the seat in front of you.", "start": 0.0, "end": 23.2, "chunk_index": 0, "language": "Chinese", "finish_reason": "eos", "truncated": false, "generated_tokens": 72, "max_new_tokens": 311 } ], "detected_language": "Chinese", "segments": [ { "start": 177.7, "end": 200.89999999999998, "text": "ご到着ありがとうございます。Welcome aboard, A N A。これより機内の安全についてご案内します。May we have your attention for the following safety instructions?荷物は座席の上の物入れまたは前の座席の下に収納してください。Please place your baggage in the overhead compartment or under the seat in front of you." } ], "start": 177.7, "end": 200.9, "engine": "mlx-qwen3-asr", "language": "zh" } }, "raw_word_timestamps": [ { "word": "請", "start": 182.88, "end": 183.42, "probability": 0.013397216796875 }, { "word": "介", "start": 183.42, "end": 183.42, "probability": 0.1248779296875 }, { "word": "紹", "start": 183.42, "end": 183.57999999999998, "probability": 0.8935546875 }, { "word": "一下", "start": 183.57999999999998, "end": 183.78, "probability": 0.146728515625 }, { "word": "機", "start": 183.78, "end": 183.78, "probability": 0.34326171875 }, { "word": "的", "start": 183.78, "end": 184.01999999999998, "probability": 0.1142578125 }, { "word": "安全", "start": 184.01999999999998, "end": 184.38, "probability": 0.9921875 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "請介紹機內的安全", "review_status": "BEST_SUPPORTED_WITH_REVIEW", "classification": "BEST_SUPPORTED_WITH_REVIEW", "confidence": 0.85, "correction_reason": "Context '機内的安全' and rejected alternative support '機內' over '機'." }, { "start": 184.38, "end": 188.7, "text": "我們請您注意安全的指示", "raw_asr_text": "我們請你注意安全的指示", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 11, "words": [], "asr_metrics": { "avg_logprob": -0.4212241617838542, "compression_ratio": 4.777777777777778, "no_speech_prob": 3.221098041072956e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "zh", "text": "謝謝你搭乘的歡迎來到A&A請介紹一下機的安全我們請你注意安全的指示請坐坐上的物入或是坐坐下的坐坐下拜拜" }, { "language": "en", "text": "Thank you for your ride. Welcome aboard A&A. I will tell you about the safety of the car. Here we have your attention for the following safety instructions. Your luggage is on the seat of the seat. Please put your luggage in the overhead compartment or under the seat in front of you." } ], "id": 32, "review_annotation": "待確認", "row_id": 32, "bounded_row_id": 32, "parent_row_ids": [ 32 ], "parent_row_id": 32, "packet_id": 11, "LID_evidence": [ { "region_id": 28, "probabilities": { "ja": 0.14047966427824576, "zh": 0.07362690406625916, "en": 0.3789523937230115 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 28 ], "language_confidence": 0.07362690406625916, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 177.7, "packet_end": 200.9, "raw_text": "我們請你注意安全的指示", "selected_text": "我們請您注意安全的指示", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 32 ], "packet_id": 11, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "我們請你注意安全的指示", "after": "我們請您注意安全的指示", "classification": "BEST_SUPPORTED_WITH_REVIEW", "reason": "Standard polite form '您' supported by rejected alternative and context." } ], "language": "zh", "packet_ids": [ 11 ], "index": 32, "independent_ASR": { "engine": "whisper-large-v3", "model_path": "/Users/sagawa/AI_WORK/cache/huggingface/hub/models--mlx-community--whisper-large-v3", "language": "zh", "audio_interval": [ 177.7, 200.9 ], "target_interval": [ 178.5, 200.1 ], "text": "请不吝点赞订阅转发打赏支持明镜与点点栏目", "words": [ { "word": "请", "start": 180.26, "end": 180.85999999999999, "probability": 0.343994140625 }, { "word": "不", "start": 180.85999999999999, "end": 180.85999999999999, "probability": 0.0040740966796875 }, { "word": "吝", "start": 180.85999999999999, "end": 183.45999999999998, "probability": 0.98828125 }, { "word": "点", "start": 183.45999999999998, "end": 183.61999999999998, "probability": 0.9990234375 }, { "word": "赞", "start": 183.61999999999998, "end": 183.66, "probability": 0.9990234375 }, { "word": " 订", "start": 183.66, "end": 183.83999999999997, "probability": 0.9990234375 }, { "word": "阅", "start": 183.83999999999997, "end": 184.0, "probability": 0.99951171875 }, { "word": " 转", "start": 184.0, "end": 184.38, "probability": 0.9990234375 }, { "word": "发", "start": 184.38, "end": 184.38, "probability": 0.99658203125 }, { "word": " 打", "start": 184.45999999999998, "end": 184.56, "probability": 0.9990234375 }, { "word": "赏", "start": 184.56, "end": 191.76, "probability": 0.9990234375 }, { "word": "支持", "start": 191.76, "end": 193.17999999999998, "probability": 0.99951171875 }, { "word": "明", "start": 193.17999999999998, "end": 196.98, "probability": 0.99951171875 }, { "word": "镜", "start": 196.98, "end": 197.95999999999998, "probability": 1.0 }, { "word": "与", "start": 197.95999999999998, "end": 198.23999999999998, "probability": 0.99951171875 }, { "word": "点", "start": 198.23999999999998, "end": 198.54, "probability": 0.99951171875 }, { "word": "点", "start": 198.54, "end": 198.83999999999997, "probability": 0.99951171875 }, { "word": "栏", "start": 198.83999999999997, "end": 199.07999999999998, "probability": 0.99951171875 }, { "word": "目", "start": 199.07999999999998, "end": 200.88, "probability": 1.0 } ], "raw_result": { "text": "请不吝点赞 订阅 转发 打赏支持明镜与点点栏目", "segments": [ { "id": 0, "seek": 0, "start": 2.5599999999999987, "end": 23.18, "text": "请不吝点赞 订阅 转发 打赏支持明镜与点点栏目", "tokens": [ 50365, 27908, 1960, 2392, 251, 12579, 5266, 252, 220, 7422, 95, 10034, 227, 220, 17819, 105, 28926, 220, 12467, 5266, 237, 35488, 11100, 12373, 250, 940, 236, 12579, 12579, 8987, 237, 11386, 51864 ], "temperature": 0.0, "avg_logprob": -0.1714823806987089, "compression_ratio": 0.8513513513513513, "no_speech_prob": 0.04063153639435768, "words": [ { "word": "请", "start": 2.5599999999999987, "end": 3.16, "probability": 0.343994140625 }, { "word": "不", "start": 3.16, "end": 3.16, "probability": 0.0040740966796875 }, { "word": "吝", "start": 3.16, "end": 5.76, "probability": 0.98828125 }, { "word": "点", "start": 5.76, "end": 5.92, "probability": 0.9990234375 }, { "word": "赞", "start": 5.92, "end": 5.96, "probability": 0.9990234375 }, { "word": " 订", "start": 5.96, "end": 6.14, "probability": 0.9990234375 }, { "word": "阅", "start": 6.14, "end": 6.3, "probability": 0.99951171875 }, { "word": " 转", "start": 6.3, "end": 6.68, "probability": 0.9990234375 }, { "word": "发", "start": 6.68, "end": 6.68, "probability": 0.99658203125 }, { "word": " 打", "start": 6.76, "end": 6.86, "probability": 0.9990234375 }, { "word": "赏", "start": 6.86, "end": 14.06, "probability": 0.9990234375 }, { "word": "支持", "start": 14.06, "end": 15.48, "probability": 0.99951171875 }, { "word": "明", "start": 15.48, "end": 19.28, "probability": 0.99951171875 }, { "word": "镜", "start": 19.28, "end": 20.26, "probability": 1.0 }, { "word": "与", "start": 20.26, "end": 20.54, "probability": 0.99951171875 }, { "word": "点", "start": 20.54, "end": 20.84, "probability": 0.99951171875 }, { "word": "点", "start": 20.84, "end": 21.14, "probability": 0.99951171875 }, { "word": "栏", "start": 21.14, "end": 21.38, "probability": 0.99951171875 }, { "word": "目", "start": 21.38, "end": 23.18, "probability": 1.0 } ] } ], "language": "zh", "duration": 23.2 }, "status": "COMPLETE", "rejected_alternative": { "text": "ご到着ありがとうございます。Welcome aboard, A N A。これより機内の安全についてご案内します。May we have your attention for the following safety instructions?荷物は座席の上の物入れまたは前の座席の下に収納してください。Please place your baggage in the overhead compartment or under the seat in front of you.", "chunks": [ { "text": "ご到着ありがとうございます。Welcome aboard, A N A。これより機内の安全についてご案内します。May we have your attention for the following safety instructions?荷物は座席の上の物入れまたは前の座席の下に収納してください。Please place your baggage in the overhead compartment or under the seat in front of you.", "start": 0.0, "end": 23.2, "chunk_index": 0, "language": "Chinese", "finish_reason": "eos", "truncated": false, "generated_tokens": 72, "max_new_tokens": 311 } ], "detected_language": "Chinese", "segments": [ { "start": 177.7, "end": 200.89999999999998, "text": "ご到着ありがとうございます。Welcome aboard, A N A。これより機内の安全についてご案内します。May we have your attention for the following safety instructions?荷物は座席の上の物入れまたは前の座席の下に収納してください。Please place your baggage in the overhead compartment or under the seat in front of you." } ], "start": 177.7, "end": 200.9, "engine": "mlx-qwen3-asr", "language": "zh" } }, "raw_word_timestamps": [ { "word": "我們", "start": 184.38, "end": 185.11999999999998, "probability": 0.0030803680419921875 }, { "word": "請", "start": 185.11999999999998, "end": 186.66, "probability": 0.107421875 }, { "word": "你", "start": 186.66, "end": 186.66, "probability": 0.056732177734375 }, { "word": "注意", "start": 186.66, "end": 186.88, "probability": 0.0156402587890625 }, { "word": "安全", "start": 186.88, "end": 187.89999999999998, "probability": 0.151123046875 }, { "word": "的", "start": 187.89999999999998, "end": 187.89999999999998, "probability": 0.11749267578125 }, { "word": "指", "start": 187.89999999999998, "end": 187.89999999999998, "probability": 0.09771728515625 }, { "word": "示", "start": 187.89999999999998, "end": 188.7, "probability": 0.4091796875 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "我們請您注意安全的指示", "review_status": "BEST_SUPPORTED_WITH_REVIEW", "classification": "BEST_SUPPORTED_WITH_REVIEW", "confidence": 0.85, "correction_reason": "Standard polite form '您' supported by rejected alternative and context." }, { "start": 188.7, "end": 191.7, "text": "請將行李放入上方的物入", "raw_asr_text": "請坐坐上的物入", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 11, "words": [], "asr_metrics": { "avg_logprob": -0.4212241617838542, "compression_ratio": 4.777777777777778, "no_speech_prob": 3.221098041072956e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "zh", "text": "謝謝你搭乘的歡迎來到A&A請介紹一下機的安全我們請你注意安全的指示請坐坐上的物入或是坐坐下的坐坐下拜拜" }, { "language": "en", "text": "Thank you for your ride. Welcome aboard A&A. I will tell you about the safety of the car. Here we have your attention for the following safety instructions. Your luggage is on the seat of the seat. Please put your luggage in the overhead compartment or under the seat in front of you." } ], "id": 33, "review_annotation": "待確認", "row_id": 33, "bounded_row_id": 33, "parent_row_ids": [ 33 ], "parent_row_id": 33, "packet_id": 11, "LID_evidence": [ { "region_id": 28, "probabilities": { "ja": 0.14047966427824576, "zh": 0.07362690406625916, "en": 0.3789523937230115 }, "status": "UNCERTAIN" }, { "region_id": 29, "probabilities": { "ja": 0.15268940564973968, "zh": 0.0033992218668572605, "en": 0.40580241428965347 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 28, 29 ], "language_confidence": 0.07362690406625916, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 177.7, "packet_end": 200.9, "raw_text": "請坐坐上的物入", "selected_text": "請將行李放入上方的物入", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 33 ], "packet_id": 11, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "請坐坐上的物入", "after": "請將行李放入上方的物入", "classification": "BEST_SUPPORTED_WITH_REVIEW", "reason": "Context '物入れ' and rejected alternative support '將行李放入' over '坐坐上的'." } ], "language": "zh", "packet_ids": [ 11 ], "index": 33, "independent_ASR": { "engine": "whisper-large-v3", "model_path": "/Users/sagawa/AI_WORK/cache/huggingface/hub/models--mlx-community--whisper-large-v3", "language": "zh", "audio_interval": [ 177.7, 200.9 ], "target_interval": [ 178.5, 200.1 ], "text": "请不吝点赞订阅转发打赏支持明镜与点点栏目", "words": [ { "word": "请", "start": 180.26, "end": 180.85999999999999, "probability": 0.343994140625 }, { "word": "不", "start": 180.85999999999999, "end": 180.85999999999999, "probability": 0.0040740966796875 }, { "word": "吝", "start": 180.85999999999999, "end": 183.45999999999998, "probability": 0.98828125 }, { "word": "点", "start": 183.45999999999998, "end": 183.61999999999998, "probability": 0.9990234375 }, { "word": "赞", "start": 183.61999999999998, "end": 183.66, "probability": 0.9990234375 }, { "word": " 订", "start": 183.66, "end": 183.83999999999997, "probability": 0.9990234375 }, { "word": "阅", "start": 183.83999999999997, "end": 184.0, "probability": 0.99951171875 }, { "word": " 转", "start": 184.0, "end": 184.38, "probability": 0.9990234375 }, { "word": "发", "start": 184.38, "end": 184.38, "probability": 0.99658203125 }, { "word": " 打", "start": 184.45999999999998, "end": 184.56, "probability": 0.9990234375 }, { "word": "赏", "start": 184.56, "end": 191.76, "probability": 0.9990234375 }, { "word": "支持", "start": 191.76, "end": 193.17999999999998, "probability": 0.99951171875 }, { "word": "明", "start": 193.17999999999998, "end": 196.98, "probability": 0.99951171875 }, { "word": "镜", "start": 196.98, "end": 197.95999999999998, "probability": 1.0 }, { "word": "与", "start": 197.95999999999998, "end": 198.23999999999998, "probability": 0.99951171875 }, { "word": "点", "start": 198.23999999999998, "end": 198.54, "probability": 0.99951171875 }, { "word": "点", "start": 198.54, "end": 198.83999999999997, "probability": 0.99951171875 }, { "word": "栏", "start": 198.83999999999997, "end": 199.07999999999998, "probability": 0.99951171875 }, { "word": "目", "start": 199.07999999999998, "end": 200.88, "probability": 1.0 } ], "raw_result": { "text": "请不吝点赞 订阅 转发 打赏支持明镜与点点栏目", "segments": [ { "id": 0, "seek": 0, "start": 2.5599999999999987, "end": 23.18, "text": "请不吝点赞 订阅 转发 打赏支持明镜与点点栏目", "tokens": [ 50365, 27908, 1960, 2392, 251, 12579, 5266, 252, 220, 7422, 95, 10034, 227, 220, 17819, 105, 28926, 220, 12467, 5266, 237, 35488, 11100, 12373, 250, 940, 236, 12579, 12579, 8987, 237, 11386, 51864 ], "temperature": 0.0, "avg_logprob": -0.1714823806987089, "compression_ratio": 0.8513513513513513, "no_speech_prob": 0.04063153639435768, "words": [ { "word": "请", "start": 2.5599999999999987, "end": 3.16, "probability": 0.343994140625 }, { "word": "不", "start": 3.16, "end": 3.16, "probability": 0.0040740966796875 }, { "word": "吝", "start": 3.16, "end": 5.76, "probability": 0.98828125 }, { "word": "点", "start": 5.76, "end": 5.92, "probability": 0.9990234375 }, { "word": "赞", "start": 5.92, "end": 5.96, "probability": 0.9990234375 }, { "word": " 订", "start": 5.96, "end": 6.14, "probability": 0.9990234375 }, { "word": "阅", "start": 6.14, "end": 6.3, "probability": 0.99951171875 }, { "word": " 转", "start": 6.3, "end": 6.68, "probability": 0.9990234375 }, { "word": "发", "start": 6.68, "end": 6.68, "probability": 0.99658203125 }, { "word": " 打", "start": 6.76, "end": 6.86, "probability": 0.9990234375 }, { "word": "赏", "start": 6.86, "end": 14.06, "probability": 0.9990234375 }, { "word": "支持", "start": 14.06, "end": 15.48, "probability": 0.99951171875 }, { "word": "明", "start": 15.48, "end": 19.28, "probability": 0.99951171875 }, { "word": "镜", "start": 19.28, "end": 20.26, "probability": 1.0 }, { "word": "与", "start": 20.26, "end": 20.54, "probability": 0.99951171875 }, { "word": "点", "start": 20.54, "end": 20.84, "probability": 0.99951171875 }, { "word": "点", "start": 20.84, "end": 21.14, "probability": 0.99951171875 }, { "word": "栏", "start": 21.14, "end": 21.38, "probability": 0.99951171875 }, { "word": "目", "start": 21.38, "end": 23.18, "probability": 1.0 } ] } ], "language": "zh", "duration": 23.2 }, "status": "COMPLETE", "rejected_alternative": { "text": "ご到着ありがとうございます。Welcome aboard, A N A。これより機内の安全についてご案内します。May we have your attention for the following safety instructions?荷物は座席の上の物入れまたは前の座席の下に収納してください。Please place your baggage in the overhead compartment or under the seat in front of you.", "chunks": [ { "text": "ご到着ありがとうございます。Welcome aboard, A N A。これより機内の安全についてご案内します。May we have your attention for the following safety instructions?荷物は座席の上の物入れまたは前の座席の下に収納してください。Please place your baggage in the overhead compartment or under the seat in front of you.", "start": 0.0, "end": 23.2, "chunk_index": 0, "language": "Chinese", "finish_reason": "eos", "truncated": false, "generated_tokens": 72, "max_new_tokens": 311 } ], "detected_language": "Chinese", "segments": [ { "start": 177.7, "end": 200.89999999999998, "text": "ご到着ありがとうございます。Welcome aboard, A N A。これより機内の安全についてご案内します。May we have your attention for the following safety instructions?荷物は座席の上の物入れまたは前の座席の下に収納してください。Please place your baggage in the overhead compartment or under the seat in front of you." } ], "start": 177.7, "end": 200.9, "engine": "mlx-qwen3-asr", "language": "zh" } }, "raw_word_timestamps": [ { "word": "請", "start": 188.7, "end": 188.95999999999998, "probability": 0.01348876953125 }, { "word": "坐", "start": 188.95999999999998, "end": 190.14, "probability": 0.12371826171875 }, { "word": "坐", "start": 190.14, "end": 190.73999999999998, "probability": 0.12646484375 }, { "word": "上", "start": 190.73999999999998, "end": 191.06, "probability": 0.266357421875 }, { "word": "的", "start": 191.06, "end": 191.29999999999998, "probability": 0.335693359375 }, { "word": "物", "start": 191.29999999999998, "end": 191.45999999999998, "probability": 0.208984375 }, { "word": "入", "start": 191.45999999999998, "end": 191.7, "probability": 0.29052734375 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "請將行李放入上方的物入", "review_status": "BEST_SUPPORTED_WITH_REVIEW", "classification": "BEST_SUPPORTED_WITH_REVIEW", "confidence": 0.8, "correction_reason": "Context '物入れ' and rejected alternative support '將行李放入' over '坐坐上的'." }, { "start": 191.7, "end": 194.7, "text": "或是坐下的行李架", "raw_asr_text": "或是坐坐下的坐坐下", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 11, "words": [], "asr_metrics": { "avg_logprob": -0.4212241617838542, "compression_ratio": 4.777777777777778, "no_speech_prob": 3.221098041072956e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "zh", "text": "謝謝你搭乘的歡迎來到A&A請介紹一下機的安全我們請你注意安全的指示請坐坐上的物入或是坐坐下的坐坐下拜拜" }, { "language": "en", "text": "Thank you for your ride. Welcome aboard A&A. I will tell you about the safety of the car. Here we have your attention for the following safety instructions. Your luggage is on the seat of the seat. Please put your luggage in the overhead compartment or under the seat in front of you." } ], "id": 34, "review_annotation": "待確認", "row_id": 34, "bounded_row_id": 34, "parent_row_ids": [ 34 ], "parent_row_id": 34, "packet_id": 11, "LID_evidence": [ { "region_id": 29, "probabilities": { "ja": 0.15268940564973968, "zh": 0.0033992218668572605, "en": 0.40580241428965347 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 29 ], "language_confidence": 0.0033992218668572605, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 177.7, "packet_end": 200.9, "raw_text": "或是坐坐下的坐坐下", "selected_text": "或是坐下的行李架", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 34 ], "packet_id": 11, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "或是坐坐下的坐坐下", "after": "或是坐下的行李架", "classification": "BEST_SUPPORTED_WITH_REVIEW", "reason": "Context '前の座席の下' and rejected alternative support '行李架' (rack) over nonsense." } ], "language": "zh", "packet_ids": [ 11 ], "index": 34, "independent_ASR": { "engine": "whisper-large-v3", "model_path": "/Users/sagawa/AI_WORK/cache/huggingface/hub/models--mlx-community--whisper-large-v3", "language": "zh", "audio_interval": [ 177.7, 200.9 ], "target_interval": [ 178.5, 200.1 ], "text": "请不吝点赞订阅转发打赏支持明镜与点点栏目", "words": [ { "word": "请", "start": 180.26, "end": 180.85999999999999, "probability": 0.343994140625 }, { "word": "不", "start": 180.85999999999999, "end": 180.85999999999999, "probability": 0.0040740966796875 }, { "word": "吝", "start": 180.85999999999999, "end": 183.45999999999998, "probability": 0.98828125 }, { "word": "点", "start": 183.45999999999998, "end": 183.61999999999998, "probability": 0.9990234375 }, { "word": "赞", "start": 183.61999999999998, "end": 183.66, "probability": 0.9990234375 }, { "word": " 订", "start": 183.66, "end": 183.83999999999997, "probability": 0.9990234375 }, { "word": "阅", "start": 183.83999999999997, "end": 184.0, "probability": 0.99951171875 }, { "word": " 转", "start": 184.0, "end": 184.38, "probability": 0.9990234375 }, { "word": "发", "start": 184.38, "end": 184.38, "probability": 0.99658203125 }, { "word": " 打", "start": 184.45999999999998, "end": 184.56, "probability": 0.9990234375 }, { "word": "赏", "start": 184.56, "end": 191.76, "probability": 0.9990234375 }, { "word": "支持", "start": 191.76, "end": 193.17999999999998, "probability": 0.99951171875 }, { "word": "明", "start": 193.17999999999998, "end": 196.98, "probability": 0.99951171875 }, { "word": "镜", "start": 196.98, "end": 197.95999999999998, "probability": 1.0 }, { "word": "与", "start": 197.95999999999998, "end": 198.23999999999998, "probability": 0.99951171875 }, { "word": "点", "start": 198.23999999999998, "end": 198.54, "probability": 0.99951171875 }, { "word": "点", "start": 198.54, "end": 198.83999999999997, "probability": 0.99951171875 }, { "word": "栏", "start": 198.83999999999997, "end": 199.07999999999998, "probability": 0.99951171875 }, { "word": "目", "start": 199.07999999999998, "end": 200.88, "probability": 1.0 } ], "raw_result": { "text": "请不吝点赞 订阅 转发 打赏支持明镜与点点栏目", "segments": [ { "id": 0, "seek": 0, "start": 2.5599999999999987, "end": 23.18, "text": "请不吝点赞 订阅 转发 打赏支持明镜与点点栏目", "tokens": [ 50365, 27908, 1960, 2392, 251, 12579, 5266, 252, 220, 7422, 95, 10034, 227, 220, 17819, 105, 28926, 220, 12467, 5266, 237, 35488, 11100, 12373, 250, 940, 236, 12579, 12579, 8987, 237, 11386, 51864 ], "temperature": 0.0, "avg_logprob": -0.1714823806987089, "compression_ratio": 0.8513513513513513, "no_speech_prob": 0.04063153639435768, "words": [ { "word": "请", "start": 2.5599999999999987, "end": 3.16, "probability": 0.343994140625 }, { "word": "不", "start": 3.16, "end": 3.16, "probability": 0.0040740966796875 }, { "word": "吝", "start": 3.16, "end": 5.76, "probability": 0.98828125 }, { "word": "点", "start": 5.76, "end": 5.92, "probability": 0.9990234375 }, { "word": "赞", "start": 5.92, "end": 5.96, "probability": 0.9990234375 }, { "word": " 订", "start": 5.96, "end": 6.14, "probability": 0.9990234375 }, { "word": "阅", "start": 6.14, "end": 6.3, "probability": 0.99951171875 }, { "word": " 转", "start": 6.3, "end": 6.68, "probability": 0.9990234375 }, { "word": "发", "start": 6.68, "end": 6.68, "probability": 0.99658203125 }, { "word": " 打", "start": 6.76, "end": 6.86, "probability": 0.9990234375 }, { "word": "赏", "start": 6.86, "end": 14.06, "probability": 0.9990234375 }, { "word": "支持", "start": 14.06, "end": 15.48, "probability": 0.99951171875 }, { "word": "明", "start": 15.48, "end": 19.28, "probability": 0.99951171875 }, { "word": "镜", "start": 19.28, "end": 20.26, "probability": 1.0 }, { "word": "与", "start": 20.26, "end": 20.54, "probability": 0.99951171875 }, { "word": "点", "start": 20.54, "end": 20.84, "probability": 0.99951171875 }, { "word": "点", "start": 20.84, "end": 21.14, "probability": 0.99951171875 }, { "word": "栏", "start": 21.14, "end": 21.38, "probability": 0.99951171875 }, { "word": "目", "start": 21.38, "end": 23.18, "probability": 1.0 } ] } ], "language": "zh", "duration": 23.2 }, "status": "COMPLETE", "rejected_alternative": { "text": "ご到着ありがとうございます。Welcome aboard, A N A。これより機内の安全についてご案内します。May we have your attention for the following safety instructions?荷物は座席の上の物入れまたは前の座席の下に収納してください。Please place your baggage in the overhead compartment or under the seat in front of you.", "chunks": [ { "text": "ご到着ありがとうございます。Welcome aboard, A N A。これより機内の安全についてご案内します。May we have your attention for the following safety instructions?荷物は座席の上の物入れまたは前の座席の下に収納してください。Please place your baggage in the overhead compartment or under the seat in front of you.", "start": 0.0, "end": 23.2, "chunk_index": 0, "language": "Chinese", "finish_reason": "eos", "truncated": false, "generated_tokens": 72, "max_new_tokens": 311 } ], "detected_language": "Chinese", "segments": [ { "start": 177.7, "end": 200.89999999999998, "text": "ご到着ありがとうございます。Welcome aboard, A N A。これより機内の安全についてご案内します。May we have your attention for the following safety instructions?荷物は座席の上の物入れまたは前の座席の下に収納してください。Please place your baggage in the overhead compartment or under the seat in front of you." } ], "start": 177.7, "end": 200.9, "engine": "mlx-qwen3-asr", "language": "zh" } }, "raw_word_timestamps": [ { "word": "或", "start": 191.7, "end": 192.2, "probability": 0.103271484375 }, { "word": "是", "start": 192.2, "end": 192.39999999999998, "probability": 0.36328125 }, { "word": "坐", "start": 192.39999999999998, "end": 193.06, "probability": 0.1605224609375 }, { "word": "坐", "start": 193.06, "end": 193.2, "probability": 0.2607421875 }, { "word": "下", "start": 193.2, "end": 193.64, "probability": 0.72021484375 }, { "word": "的", "start": 193.64, "end": 193.83999999999997, "probability": 0.79638671875 }, { "word": "坐", "start": 193.83999999999997, "end": 193.98, "probability": 0.10565185546875 }, { "word": "坐", "start": 193.98, "end": 194.16, "probability": 0.352294921875 }, { "word": "下", "start": 194.16, "end": 194.7, "probability": 0.411376953125 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "或是坐下的行李架", "review_status": "BEST_SUPPORTED_WITH_REVIEW", "classification": "BEST_SUPPORTED_WITH_REVIEW", "confidence": 0.8, "correction_reason": "Context '前の座席の下' and rejected alternative support '行李架' (rack) over nonsense." }, { "start": 194.7, "end": 195.7, "text": "[聴不清／待確認]", "raw_asr_text": "拜拜", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 11, "words": [], "asr_metrics": { "avg_logprob": -1.5033599138259888, "compression_ratio": 0.42857142857142855, "no_speech_prob": 1.2041420638375655e-10 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "zh", "text": "謝謝你搭乘的歡迎來到A&A請介紹一下機的安全我們請你注意安全的指示請坐坐上的物入或是坐坐下的坐坐下拜拜" }, { "language": "en", "text": "Thank you for your ride. Welcome aboard A&A. I will tell you about the safety of the car. Here we have your attention for the following safety instructions. Your luggage is on the seat of the seat. Please put your luggage in the overhead compartment or under the seat in front of you." } ], "id": 35, "review_annotation": "待確認", "row_id": 35, "bounded_row_id": 35, "parent_row_ids": [ 35 ], "parent_row_id": 35, "packet_id": 11, "LID_evidence": [ { "region_id": 29, "probabilities": { "ja": 0.15268940564973968, "zh": 0.0033992218668572605, "en": 0.40580241428965347 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 29 ], "language_confidence": 0.0033992218668572605, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 177.7, "packet_end": 200.9, "raw_text": "拜拜", "selected_text": "[聴不清／待確認]", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 35 ], "packet_id": 11, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "拜拜", "after": "[聴不清／待確認]", "classification": "UNRECOVERABLE_MARKED", "reason": "Audio contains unrelated promotional content; no valid speech recovery exists." } ], "language": "zh", "packet_ids": [ 11 ], "index": 35, "independent_ASR": { "engine": "whisper-large-v3", "model_path": "/Users/sagawa/AI_WORK/cache/huggingface/hub/models--mlx-community--whisper-large-v3", "language": "zh", "audio_interval": [ 177.7, 200.9 ], "target_interval": [ 178.5, 200.1 ], "text": "请不吝点赞订阅转发打赏支持明镜与点点栏目", "words": [ { "word": "请", "start": 180.26, "end": 180.85999999999999, "probability": 0.343994140625 }, { "word": "不", "start": 180.85999999999999, "end": 180.85999999999999, "probability": 0.0040740966796875 }, { "word": "吝", "start": 180.85999999999999, "end": 183.45999999999998, "probability": 0.98828125 }, { "word": "点", "start": 183.45999999999998, "end": 183.61999999999998, "probability": 0.9990234375 }, { "word": "赞", "start": 183.61999999999998, "end": 183.66, "probability": 0.9990234375 }, { "word": " 订", "start": 183.66, "end": 183.83999999999997, "probability": 0.9990234375 }, { "word": "阅", "start": 183.83999999999997, "end": 184.0, "probability": 0.99951171875 }, { "word": " 转", "start": 184.0, "end": 184.38, "probability": 0.9990234375 }, { "word": "发", "start": 184.38, "end": 184.38, "probability": 0.99658203125 }, { "word": " 打", "start": 184.45999999999998, "end": 184.56, "probability": 0.9990234375 }, { "word": "赏", "start": 184.56, "end": 191.76, "probability": 0.9990234375 }, { "word": "支持", "start": 191.76, "end": 193.17999999999998, "probability": 0.99951171875 }, { "word": "明", "start": 193.17999999999998, "end": 196.98, "probability": 0.99951171875 }, { "word": "镜", "start": 196.98, "end": 197.95999999999998, "probability": 1.0 }, { "word": "与", "start": 197.95999999999998, "end": 198.23999999999998, "probability": 0.99951171875 }, { "word": "点", "start": 198.23999999999998, "end": 198.54, "probability": 0.99951171875 }, { "word": "点", "start": 198.54, "end": 198.83999999999997, "probability": 0.99951171875 }, { "word": "栏", "start": 198.83999999999997, "end": 199.07999999999998, "probability": 0.99951171875 }, { "word": "目", "start": 199.07999999999998, "end": 200.88, "probability": 1.0 } ], "raw_result": { "text": "请不吝点赞 订阅 转发 打赏支持明镜与点点栏目", "segments": [ { "id": 0, "seek": 0, "start": 2.5599999999999987, "end": 23.18, "text": "请不吝点赞 订阅 转发 打赏支持明镜与点点栏目", "tokens": [ 50365, 27908, 1960, 2392, 251, 12579, 5266, 252, 220, 7422, 95, 10034, 227, 220, 17819, 105, 28926, 220, 12467, 5266, 237, 35488, 11100, 12373, 250, 940, 236, 12579, 12579, 8987, 237, 11386, 51864 ], "temperature": 0.0, "avg_logprob": -0.1714823806987089, "compression_ratio": 0.8513513513513513, "no_speech_prob": 0.04063153639435768, "words": [ { "word": "请", "start": 2.5599999999999987, "end": 3.16, "probability": 0.343994140625 }, { "word": "不", "start": 3.16, "end": 3.16, "probability": 0.0040740966796875 }, { "word": "吝", "start": 3.16, "end": 5.76, "probability": 0.98828125 }, { "word": "点", "start": 5.76, "end": 5.92, "probability": 0.9990234375 }, { "word": "赞", "start": 5.92, "end": 5.96, "probability": 0.9990234375 }, { "word": " 订", "start": 5.96, "end": 6.14, "probability": 0.9990234375 }, { "word": "阅", "start": 6.14, "end": 6.3, "probability": 0.99951171875 }, { "word": " 转", "start": 6.3, "end": 6.68, "probability": 0.9990234375 }, { "word": "发", "start": 6.68, "end": 6.68, "probability": 0.99658203125 }, { "word": " 打", "start": 6.76, "end": 6.86, "probability": 0.9990234375 }, { "word": "赏", "start": 6.86, "end": 14.06, "probability": 0.9990234375 }, { "word": "支持", "start": 14.06, "end": 15.48, "probability": 0.99951171875 }, { "word": "明", "start": 15.48, "end": 19.28, "probability": 0.99951171875 }, { "word": "镜", "start": 19.28, "end": 20.26, "probability": 1.0 }, { "word": "与", "start": 20.26, "end": 20.54, "probability": 0.99951171875 }, { "word": "点", "start": 20.54, "end": 20.84, "probability": 0.99951171875 }, { "word": "点", "start": 20.84, "end": 21.14, "probability": 0.99951171875 }, { "word": "栏", "start": 21.14, "end": 21.38, "probability": 0.99951171875 }, { "word": "目", "start": 21.38, "end": 23.18, "probability": 1.0 } ] } ], "language": "zh", "duration": 23.2 }, "status": "COMPLETE", "rejected_alternative": { "text": "ご到着ありがとうございます。Welcome aboard, A N A。これより機内の安全についてご案内します。May we have your attention for the following safety instructions?荷物は座席の上の物入れまたは前の座席の下に収納してください。Please place your baggage in the overhead compartment or under the seat in front of you.", "chunks": [ { "text": "ご到着ありがとうございます。Welcome aboard, A N A。これより機内の安全についてご案内します。May we have your attention for the following safety instructions?荷物は座席の上の物入れまたは前の座席の下に収納してください。Please place your baggage in the overhead compartment or under the seat in front of you.", "start": 0.0, "end": 23.2, "chunk_index": 0, "language": "Chinese", "finish_reason": "eos", "truncated": false, "generated_tokens": 72, "max_new_tokens": 311 } ], "detected_language": "Chinese", "segments": [ { "start": 177.7, "end": 200.89999999999998, "text": "ご到着ありがとうございます。Welcome aboard, A N A。これより機内の安全についてご案内します。May we have your attention for the following safety instructions?荷物は座席の上の物入れまたは前の座席の下に収納してください。Please place your baggage in the overhead compartment or under the seat in front of you." } ], "start": 177.7, "end": 200.9, "engine": "mlx-qwen3-asr", "language": "zh" } }, "raw_word_timestamps": [ { "word": "拜拜", "start": 194.7, "end": 195.7, "probability": 0.07177734375 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "[聴不清／待確認]", "review_status": "UNRECOVERABLE_MARKED", "classification": "UNRECOVERABLE_MARKED", "confidence": 0.1, "correction_reason": "Audio contains unrelated promotional content; no valid speech recovery exists." }, { "start": 202.6, "end": 207.64, "text": "脱出の妨げとなるため通路や非常口付近には手荷物を置かないでください", "raw_asr_text": "脱出の妨げとなるため通路や非常口付近には手荷物を置かないでください", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 12, "words": [ { "word": "脱", "start": 202.6, "end": 202.93999999999997, "probability": 0.9931640625 }, { "word": "出", "start": 202.93999999999997, "end": 203.14, "probability": 0.99462890625 }, { "word": "の", "start": 203.14, "end": 203.29999999999998, "probability": 0.99658203125 }, { "word": "妨", "start": 203.29999999999998, "end": 203.55999999999997, "probability": 0.98974609375 }, { "word": "げ", "start": 203.55999999999997, "end": 203.71999999999997, "probability": 0.99658203125 }, { "word": "と", "start": 203.71999999999997, "end": 203.85999999999999, "probability": 0.998046875 }, { "word": "なる", "start": 203.85999999999999, "end": 204.05999999999997, "probability": 1.0 }, { "word": "ため", "start": 204.05999999999997, "end": 204.35999999999999, "probability": 0.9990234375 }, { "word": "通", "start": 204.35999999999999, "end": 204.82, "probability": 0.72509765625 }, { "word": "路", "start": 204.82, "end": 204.98, "probability": 1.0 }, { "word": "や", "start": 204.98, "end": 205.11999999999998, "probability": 0.99951171875 }, { "word": "非常", "start": 205.11999999999998, "end": 205.35999999999999, "probability": 0.99951171875 }, { "word": "口", "start": 205.35999999999999, "end": 205.57999999999998, "probability": 0.99755859375 }, { "word": "付", "start": 205.57999999999998, "end": 205.76, "probability": 0.9990234375 }, { "word": "近", "start": 205.76, "end": 205.92, "probability": 1.0 }, { "word": "には", "start": 205.92, "end": 206.16, "probability": 0.998046875 }, { "word": "手", "start": 206.16, "end": 206.38, "probability": 0.984375 }, { "word": "荷", "start": 206.38, "end": 206.48, "probability": 0.9951171875 }, { "word": "物", "start": 206.48, "end": 206.66, "probability": 0.99951171875 }, { "word": "を", "start": 206.66, "end": 206.82, "probability": 0.99951171875 }, { "word": "置", "start": 206.82, "end": 206.92, "probability": 0.99951171875 }, { "word": "か", "start": 206.92, "end": 207.01999999999998, "probability": 0.99951171875 }, { "word": "ない", "start": 207.01999999999998, "end": 207.16, "probability": 1.0 }, { "word": "で", "start": 207.16, "end": 207.27999999999997, "probability": 0.99951171875 }, { "word": "ください", "start": 207.27999999999997, "end": 207.64, "probability": 0.99755859375 } ], "asr_metrics": { "avg_logprob": -0.14500580136738125, "compression_ratio": 1.1842105263157894, "no_speech_prob": 1.638091572109257e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "脱出の妨げとなるため通路や非常口付近には手荷物を置かないでください離陸着陸の際は座席とテーブルを元の位置にお戻しくださいお客様のお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様" }, { "language": "en", "text": "For a smooth evacuation, nothing should block the aisle or exits. For a smooth evacuation, nothing should block the aisle or exits. Return your seat and table to the original position during takeoff and landing. Thank you." } ], "id": 36, "review_annotation": "", "row_id": 36, "bounded_row_id": 36, "parent_row_ids": [ 36 ], "parent_row_id": 36, "packet_id": 12, "LID_evidence": [ { "region_id": 30, "probabilities": { "ja": 0.27656045549847474, "zh": 0.003627028550909017, "en": 0.46357852723740506 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 30 ], "language_confidence": 0.27656045549847474, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 201.79999999999998, "packet_end": 224.0, "raw_text": "脱出の妨げとなるため通路や非常口付近には手荷物を置かないでください", "selected_text": "脱出の妨げとなるため通路や非常口付近には手荷物を置かないでください", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 36 ], "packet_id": 12, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "脱出の妨げとなるため通路や非常口付近には手荷物を置かないでください", "after": "脱出の妨げとなるため通路や非常口付近には手荷物を置かないでください", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard safety announcement phrasing matches context and independent ASR." } ], "language": "ja", "packet_ids": [ 12 ], "index": 36, "source_text": "脱出の妨げとなるため通路や非常口付近には手荷物を置かないでください", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Standard safety announcement phrasing matches context and independent ASR." }, { "start": 207.64, "end": 217.17999999999998, "text": "離陸着陸の際は座席とテーブルを元の位置にお戻しください", "raw_asr_text": "離陸着陸の際は座席とテーブルを元の位置にお戻しください", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 12, "words": [ { "word": "離", "start": 207.64, "end": 208.27999999999997, "probability": 0.7939453125 }, { "word": "陸", "start": 208.27999999999997, "end": 213.21999999999997, "probability": 0.9970703125 }, { "word": "着", "start": 213.21999999999997, "end": 213.55999999999997, "probability": 0.415771484375 }, { "word": "陸", "start": 213.55999999999997, "end": 213.83999999999997, "probability": 1.0 }, { "word": "の", "start": 213.83999999999997, "end": 213.98, "probability": 0.9990234375 }, { "word": "際", "start": 213.98, "end": 214.17999999999998, "probability": 1.0 }, { "word": "は", "start": 214.17999999999998, "end": 214.45999999999998, "probability": 0.99951171875 }, { "word": "座", "start": 214.45999999999998, "end": 214.77999999999997, "probability": 0.9970703125 }, { "word": "席", "start": 214.77999999999997, "end": 214.95999999999998, "probability": 1.0 }, { "word": "と", "start": 214.95999999999998, "end": 215.07999999999998, "probability": 1.0 }, { "word": "テ", "start": 215.07999999999998, "end": 215.23999999999998, "probability": 0.9990234375 }, { "word": "ーブ", "start": 215.23999999999998, "end": 215.38, "probability": 1.0 }, { "word": "ル", "start": 215.38, "end": 215.45999999999998, "probability": 1.0 }, { "word": "を", "start": 215.45999999999998, "end": 215.65999999999997, "probability": 0.99951171875 }, { "word": "元", "start": 215.65999999999997, "end": 215.88, "probability": 0.99609375 }, { "word": "の", "start": 215.88, "end": 216.04, "probability": 1.0 }, { "word": "位", "start": 216.04, "end": 216.17999999999998, "probability": 0.99951171875 }, { "word": "置", "start": 216.17999999999998, "end": 216.23999999999998, "probability": 1.0 }, { "word": "に", "start": 216.23999999999998, "end": 216.33999999999997, "probability": 1.0 }, { "word": "お", "start": 216.33999999999997, "end": 216.48, "probability": 0.95361328125 }, { "word": "戻", "start": 216.48, "end": 216.61999999999998, "probability": 0.9990234375 }, { "word": "し", "start": 216.61999999999998, "end": 216.71999999999997, "probability": 0.9990234375 }, { "word": "ください", "start": 216.71999999999997, "end": 217.17999999999998, "probability": 0.9990234375 } ], "asr_metrics": { "avg_logprob": -0.14500580136738125, "compression_ratio": 1.1842105263157894, "no_speech_prob": 1.638091572109257e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "脱出の妨げとなるため通路や非常口付近には手荷物を置かないでください離陸着陸の際は座席とテーブルを元の位置にお戻しくださいお客様のお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様" }, { "language": "en", "text": "For a smooth evacuation, nothing should block the aisle or exits. For a smooth evacuation, nothing should block the aisle or exits. Return your seat and table to the original position during takeoff and landing. Thank you." } ], "id": 37, "review_annotation": "", "row_id": 37, "bounded_row_id": 37, "parent_row_ids": [ 37 ], "parent_row_id": 37, "packet_id": 12, "LID_evidence": [ { "region_id": 30, "probabilities": { "ja": 0.27656045549847474, "zh": 0.003627028550909017, "en": 0.46357852723740506 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 30 ], "language_confidence": 0.27656045549847474, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 201.79999999999998, "packet_end": 224.0, "raw_text": "離陸着陸の際は座席とテーブルを元の位置にお戻しください", "selected_text": "離陸着陸の際は座席とテーブルを元の位置にお戻しください", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 37 ], "packet_id": 12, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "離陸着陸の際は座席とテーブルを元の位置にお戻しください", "after": "離陸着陸の際は座席とテーブルを元の位置にお戻しください", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard safety announcement phrasing matches context and independent ASR." } ], "language": "ja", "packet_ids": [ 12 ], "index": 37, "source_text": "離陸着陸の際は座席とテーブルを元の位置にお戻しください", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Standard safety announcement phrasing matches context and independent ASR." }, { "start": 217.17999999999998, "end": 223.2, "text": "Return your seat and table to the original position during takeoff and landing.", "raw_asr_text": "お客様のお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様", "source_language": "en", "source_lang": "en", "asr_packet_id": 12, "words": [], "asr_metrics": { "avg_logprob": -0.10137613084581164, "compression_ratio": 21.580645161290324, "no_speech_prob": 2.2980678124340237e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "脱出の妨げとなるため通路や非常口付近には手荷物を置かないでください離陸着陸の際は座席とテーブルを元の位置にお戻しくださいお客様のお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様" }, { "language": "en", "text": "For a smooth evacuation, nothing should block the aisle or exits. For a smooth evacuation, nothing should block the aisle or exits. Return your seat and table to the original position during takeoff and landing. Thank you." } ], "id": 38, "review_annotation": "", "row_id": 38, "bounded_row_id": 38, "parent_row_ids": [ 38 ], "parent_row_id": 38, "packet_id": 12, "LID_evidence": [ { "region_id": 30, "probabilities": { "ja": 0.27656045549847474, "zh": 0.003627028550909017, "en": 0.46357852723740506 }, "status": "UNCERTAIN" }, { "region_id": 31, "probabilities": { "ja": 0.06105451409772892, "zh": 0.00402700992951092, "en": 0.42833656584843993 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 30, 31 ], "language_confidence": 0.65, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 201.79999999999998, "packet_end": 224.0, "raw_text": "お客様のお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様", "selected_text": "Return your seat and table to the original position during takeoff and landing.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 38 ], "packet_id": 12, "engine": "mlx-whisper" }, { "stage": "selector_local_rescue", "before": "お客様のお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様にお客様", "after": "Return your seat and table to the original position during takeoff and landing.", "span": [ 217.17999999999998, 223.2 ], "caption_evidence": [ { "start": 216.239, "end": 222.83, "text": "にお戻し", "source": "source_caption" }, { "start": 222.84, "end": 226.56, "text": "くださいベルト着用サイン点灯中並びにお", "source": "source_caption" } ], "audio_evidence": { "engine": "whisper-large-v3", "model_path": "/Users/sagawa/AI_WORK/cache/huggingface/hub/models--mlx-community--whisper-large-v3", "language": "en", "audio_interval": [ 217.17999999999998, 223.2 ], "target_interval": [ 217.17999999999998, 223.2 ], "text": "Return your seat and table to the original position during takeoff and landing.", "words": [ { "word": " Return", "start": 217.17999999999998, "end": 217.61999999999998, "probability": 0.92822265625 }, { "word": " your", "start": 217.61999999999998, "end": 217.98, "probability": 1.0 }, { "word": " seat", "start": 217.98, "end": 218.21999999999997, "probability": 1.0 }, { "word": " and", "start": 218.21999999999997, "end": 218.42, "probability": 1.0 }, { "word": " table", "start": 218.42, "end": 218.76, "probability": 1.0 }, { "word": " to", "start": 218.76, "end": 218.99999999999997, "probability": 1.0 }, { "word": " the", "start": 218.99999999999997, "end": 219.1, "probability": 0.99951171875 }, { "word": " original", "start": 219.1, "end": 219.45999999999998, "probability": 1.0 }, { "word": " position", "start": 219.45999999999998, "end": 219.92, "probability": 1.0 }, { "word": " during", "start": 219.92, "end": 220.48, "probability": 1.0 }, { "word": " takeoff", "start": 220.48, "end": 221.01999999999998, "probability": 0.8779296875 }, { "word": " and", "start": 221.01999999999998, "end": 221.23999999999998, "probability": 1.0 }, { "word": " landing.", "start": 221.23999999999998, "end": 221.54, "probability": 1.0 } ], "raw_result": { "text": "脱出の妨げとなるため、通路や非常口付近には手荷物を置かないでください。 For a smooth evacuation, nothing should block the aisle or exits. 離陸・着陸の際は、座席とテーブルを元の位置にお戻しください。 Return your seat and table to the original position during takeoff and landing.", "segments": [ { "id": 0, "seek": 0, "start": 5.36, "end": 6.04, "text": " 脱出の妨げとなるため、通路や非常口付近には手荷物を置かないでください。", "tokens": [ 50365, 220, 27067, 109, 7781, 2972, 5648, 101, 26012, 3193, 29218, 49983, 1231, 19550, 24658, 7355, 14392, 18144, 41469, 17463, 34779, 11389, 31409, 115, 23516, 5998, 34719, 3703, 9311, 2474, 25079, 1543, 50675 ], "temperature": 0.0, "avg_logprob": -0.13845066977019357, "compression_ratio": 1.1958041958041958, "no_speech_prob": 0.0849790871143341, "words": [ { "word": " 脱出の妨げとなるため、通路や非常口付近には手荷物を置かないでください。", "start": 5.36, "end": 6.04, "probability": 0.9423828125 } ] }, { "id": 1, "seek": 0, "start": 6.04, "end": 10.1, "text": " For a smooth evacuation, nothing should block the aisle or exits.", "tokens": [ 50675, 1171, 257, 5508, 42740, 11, 1825, 820, 3461, 264, 30916, 420, 44183, 13, 50884 ], "temperature": 0.0, "avg_logprob": -0.13845066977019357, "compression_ratio": 1.1958041958041958, "no_speech_prob": 0.0849790871143341, "words": [ { "word": " For", "start": 6.04, "end": 6.28, "probability": 0.57421875 }, { "word": " a", "start": 6.28, "end": 6.52, "probability": 0.99609375 }, { "word": " smooth", "start": 6.52, "end": 6.82, "probability": 1.0 }, { "word": " evacuation,", "start": 6.82, "end": 7.18, "probability": 0.99951171875 }, { "word": " nothing", "start": 7.92, "end": 8.34, "probability": 0.99951171875 }, { "word": " should", "start": 8.34, "end": 8.62, "probability": 1.0 }, { "word": " block", "start": 8.62, "end": 8.9, "probability": 1.0 }, { "word": " the", "start": 8.9, "end": 9.08, "probability": 1.0 }, { "word": " aisle", "start": 9.08, "end": 9.46, "probability": 0.9921875 }, { "word": " or", "start": 9.46, "end": 9.78, "probability": 0.99951171875 }, { "word": " exits.", "start": 9.78, "end": 10.1, "probability": 0.99951171875 } ] }, { "id": 2, "seek": 0, "start": 14.700000000000001, "end": 15.38, "text": " 離陸・着陸の際は、座席とテーブルを元の位置にお戻しください。", "tokens": [ 50919, 220, 32662, 8842, 116, 5379, 20708, 8842, 116, 2972, 34837, 3065, 1231, 6346, 100, 4845, 255, 3193, 22985, 15266, 244, 9405, 5998, 14812, 2972, 11160, 34719, 4108, 6117, 1486, 119, 2849, 25079, 1543, 51144 ], "temperature": 0.0, "avg_logprob": -0.13845066977019357, "compression_ratio": 1.1958041958041958, "no_speech_prob": 0.0849790871143341, "words": [ { "word": " 離陸・着陸の際は、座席とテーブルを元の位置にお戻しください。", "start": 14.700000000000001, "end": 15.38, "probability": 0.98095703125 } ] }, { "id": 3, "seek": 0, "start": 15.38, "end": 19.74, "text": " Return your seat and table to the original position during takeoff and landing.", "tokens": [ 51144, 24350, 428, 6121, 293, 3199, 281, 264, 3380, 2535, 1830, 747, 4506, 293, 11202, 13, 51359 ], "temperature": 0.0, "avg_logprob": -0.13845066977019357, "compression_ratio": 1.1958041958041958, "no_speech_prob": 0.0849790871143341, "words": [ { "word": " Return", "start": 15.38, "end": 15.82, "probability": 0.92822265625 }, { "word": " your", "start": 15.82, "end": 16.18, "probability": 1.0 }, { "word": " seat", "start": 16.18, "end": 16.42, "probability": 1.0 }, { "word": " and", "start": 16.42, "end": 16.62, "probability": 1.0 }, { "word": " table", "start": 16.62, "end": 16.96, "probability": 1.0 }, { "word": " to", "start": 16.96, "end": 17.2, "probability": 1.0 }, { "word": " the", "start": 17.2, "end": 17.3, "probability": 0.99951171875 }, { "word": " original", "start": 17.3, "end": 17.66, "probability": 1.0 }, { "word": " position", "start": 17.66, "end": 18.12, "probability": 1.0 }, { "word": " during", "start": 18.12, "end": 18.68, "probability": 1.0 }, { "word": " takeoff", "start": 18.68, "end": 19.22, "probability": 0.8779296875 }, { "word": " and", "start": 19.22, "end": 19.44, "probability": 1.0 }, { "word": " landing.", "start": 19.44, "end": 19.74, "probability": 1.0 } ] } ], "language": "en", "duration": 22.2 }, "status": "COMPLETE", "rejected_alternative": { "text": "リターン your seat and table to the original position during takeoff and landing.", "chunks": [ { "text": "リターン your seat and table to the original position during takeoff and landing.", "start": 0.0, "end": 6.02, "chunk_index": 0, "language": "Japanese", "finish_reason": "eos", "truncated": false, "generated_tokens": 17, "max_new_tokens": 128 } ], "detected_language": "Japanese", "segments": [ { "start": 217.17999999999998, "end": 223.2, "text": "リターン your seat and table to the original position during takeoff and landing." } ], "start": 217.17999999999998, "end": 223.2, "engine": "mlx-qwen3-asr", "language": "ja" }, "independent_auto_ASR": { "text": "For a smooth evacuation, nothing should block the aisle or exits. Landing.着陸の際は座席とテーブルを元の位置にお戻しください. Return your seat and table to the original position during takeoff and landing.", "chunks": [ { "text": "For a smooth evacuation, nothing should block the aisle or exits. Landing.着陸の際は座席とテーブルを元の位置にお戻しください. Return your seat and table to the original position during takeoff and landing.", "start": 0.0, "end": 22.2, "chunk_index": 0, "language": "English", "finish_reason": "eos", "truncated": false, "generated_tokens": 54, "max_new_tokens": 299 } ], "detected_language": "English", "segments": [ { "start": 201.79999999999998, "end": 223.99999999999997, "text": "For a smooth evacuation, nothing should block the aisle or exits. Landing.着陸の際は座席とテーブルを元の位置にお戻しください. Return your seat and table to the original position during takeoff and landing." } ], "start": 201.79999999999998, "end": 224.0, "engine": "mlx-qwen3-asr", "language": null }, "cross_engine_agreement": 1.0, "replacement_language": "en", "replacement": "Return your seat and table to the original position during takeoff and landing.", "decision": "REPAIRABLE" }, "language_before": "ja", "language_after": "en" }, { "stage": "canonical_review", "before": "Return your seat and table to the original position during takeoff and landing.", "after": "Return your seat and table to the original position during takeoff and landing.", "classification": "CERTAIN_RESTORATION", "reason": "Independent ASR confirms exact wording despite raw ASR failure." } ], "language": "en", "independent_ASR": { "engine": "whisper-large-v3", "model_path": "/Users/sagawa/AI_WORK/cache/huggingface/hub/models--mlx-community--whisper-large-v3", "language": "en", "audio_interval": [ 217.17999999999998, 223.2 ], "target_interval": [ 217.17999999999998, 223.2 ], "text": "Return your seat and table to the original position during takeoff and landing.", "words": [ { "word": " Return", "start": 217.17999999999998, "end": 217.61999999999998, "probability": 0.92822265625 }, { "word": " your", "start": 217.61999999999998, "end": 217.98, "probability": 1.0 }, { "word": " seat", "start": 217.98, "end": 218.21999999999997, "probability": 1.0 }, { "word": " and", "start": 218.21999999999997, "end": 218.42, "probability": 1.0 }, { "word": " table", "start": 218.42, "end": 218.76, "probability": 1.0 }, { "word": " to", "start": 218.76, "end": 218.99999999999997, "probability": 1.0 }, { "word": " the", "start": 218.99999999999997, "end": 219.1, "probability": 0.99951171875 }, { "word": " original", "start": 219.1, "end": 219.45999999999998, "probability": 1.0 }, { "word": " position", "start": 219.45999999999998, "end": 219.92, "probability": 1.0 }, { "word": " during", "start": 219.92, "end": 220.48, "probability": 1.0 }, { "word": " takeoff", "start": 220.48, "end": 221.01999999999998, "probability": 0.8779296875 }, { "word": " and", "start": 221.01999999999998, "end": 221.23999999999998, "probability": 1.0 }, { "word": " landing.", "start": 221.23999999999998, "end": 221.54, "probability": 1.0 } ], "raw_result": { "text": "脱出の妨げとなるため、通路や非常口付近には手荷物を置かないでください。 For a smooth evacuation, nothing should block the aisle or exits. 離陸・着陸の際は、座席とテーブルを元の位置にお戻しください。 Return your seat and table to the original position during takeoff and landing.", "segments": [ { "id": 0, "seek": 0, "start": 5.36, "end": 6.04, "text": " 脱出の妨げとなるため、通路や非常口付近には手荷物を置かないでください。", "tokens": [ 50365, 220, 27067, 109, 7781, 2972, 5648, 101, 26012, 3193, 29218, 49983, 1231, 19550, 24658, 7355, 14392, 18144, 41469, 17463, 34779, 11389, 31409, 115, 23516, 5998, 34719, 3703, 9311, 2474, 25079, 1543, 50675 ], "temperature": 0.0, "avg_logprob": -0.13845066977019357, "compression_ratio": 1.1958041958041958, "no_speech_prob": 0.0849790871143341, "words": [ { "word": " 脱出の妨げとなるため、通路や非常口付近には手荷物を置かないでください。", "start": 5.36, "end": 6.04, "probability": 0.9423828125 } ] }, { "id": 1, "seek": 0, "start": 6.04, "end": 10.1, "text": " For a smooth evacuation, nothing should block the aisle or exits.", "tokens": [ 50675, 1171, 257, 5508, 42740, 11, 1825, 820, 3461, 264, 30916, 420, 44183, 13, 50884 ], "temperature": 0.0, "avg_logprob": -0.13845066977019357, "compression_ratio": 1.1958041958041958, "no_speech_prob": 0.0849790871143341, "words": [ { "word": " For", "start": 6.04, "end": 6.28, "probability": 0.57421875 }, { "word": " a", "start": 6.28, "end": 6.52, "probability": 0.99609375 }, { "word": " smooth", "start": 6.52, "end": 6.82, "probability": 1.0 }, { "word": " evacuation,", "start": 6.82, "end": 7.18, "probability": 0.99951171875 }, { "word": " nothing", "start": 7.92, "end": 8.34, "probability": 0.99951171875 }, { "word": " should", "start": 8.34, "end": 8.62, "probability": 1.0 }, { "word": " block", "start": 8.62, "end": 8.9, "probability": 1.0 }, { "word": " the", "start": 8.9, "end": 9.08, "probability": 1.0 }, { "word": " aisle", "start": 9.08, "end": 9.46, "probability": 0.9921875 }, { "word": " or", "start": 9.46, "end": 9.78, "probability": 0.99951171875 }, { "word": " exits.", "start": 9.78, "end": 10.1, "probability": 0.99951171875 } ] }, { "id": 2, "seek": 0, "start": 14.700000000000001, "end": 15.38, "text": " 離陸・着陸の際は、座席とテーブルを元の位置にお戻しください。", "tokens": [ 50919, 220, 32662, 8842, 116, 5379, 20708, 8842, 116, 2972, 34837, 3065, 1231, 6346, 100, 4845, 255, 3193, 22985, 15266, 244, 9405, 5998, 14812, 2972, 11160, 34719, 4108, 6117, 1486, 119, 2849, 25079, 1543, 51144 ], "temperature": 0.0, "avg_logprob": -0.13845066977019357, "compression_ratio": 1.1958041958041958, "no_speech_prob": 0.0849790871143341, "words": [ { "word": " 離陸・着陸の際は、座席とテーブルを元の位置にお戻しください。", "start": 14.700000000000001, "end": 15.38, "probability": 0.98095703125 } ] }, { "id": 3, "seek": 0, "start": 15.38, "end": 19.74, "text": " Return your seat and table to the original position during takeoff and landing.", "tokens": [ 51144, 24350, 428, 6121, 293, 3199, 281, 264, 3380, 2535, 1830, 747, 4506, 293, 11202, 13, 51359 ], "temperature": 0.0, "avg_logprob": -0.13845066977019357, "compression_ratio": 1.1958041958041958, "no_speech_prob": 0.0849790871143341, "words": [ { "word": " Return", "start": 15.38, "end": 15.82, "probability": 0.92822265625 }, { "word": " your", "start": 15.82, "end": 16.18, "probability": 1.0 }, { "word": " seat", "start": 16.18, "end": 16.42, "probability": 1.0 }, { "word": " and", "start": 16.42, "end": 16.62, "probability": 1.0 }, { "word": " table", "start": 16.62, "end": 16.96, "probability": 1.0 }, { "word": " to", "start": 16.96, "end": 17.2, "probability": 1.0 }, { "word": " the", "start": 17.2, "end": 17.3, "probability": 0.99951171875 }, { "word": " original", "start": 17.3, "end": 17.66, "probability": 1.0 }, { "word": " position", "start": 17.66, "end": 18.12, "probability": 1.0 }, { "word": " during", "start": 18.12, "end": 18.68, "probability": 1.0 }, { "word": " takeoff", "start": 18.68, "end": 19.22, "probability": 0.8779296875 }, { "word": " and", "start": 19.22, "end": 19.44, "probability": 1.0 }, { "word": " landing.", "start": 19.44, "end": 19.74, "probability": 1.0 } ] } ], "language": "en", "duration": 22.2 }, "status": "COMPLETE", "rejected_alternative": { "text": "リターン your seat and table to the original position during takeoff and landing.", "chunks": [ { "text": "リターン your seat and table to the original position during takeoff and landing.", "start": 0.0, "end": 6.02, "chunk_index": 0, "language": "Japanese", "finish_reason": "eos", "truncated": false, "generated_tokens": 17, "max_new_tokens": 128 } ], "detected_language": "Japanese", "segments": [ { "start": 217.17999999999998, "end": 223.2, "text": "リターン your seat and table to the original position during takeoff and landing." } ], "start": 217.17999999999998, "end": 223.2, "engine": "mlx-qwen3-asr", "language": "ja" }, "independent_auto_ASR": { "text": "For a smooth evacuation, nothing should block the aisle or exits. Landing.着陸の際は座席とテーブルを元の位置にお戻しください. Return your seat and table to the original position during takeoff and landing.", "chunks": [ { "text": "For a smooth evacuation, nothing should block the aisle or exits. Landing.着陸の際は座席とテーブルを元の位置にお戻しください. Return your seat and table to the original position during takeoff and landing.", "start": 0.0, "end": 22.2, "chunk_index": 0, "language": "English", "finish_reason": "eos", "truncated": false, "generated_tokens": 54, "max_new_tokens": 299 } ], "detected_language": "English", "segments": [ { "start": 201.79999999999998, "end": 223.99999999999997, "text": "For a smooth evacuation, nothing should block the aisle or exits. Landing.着陸の際は座席とテーブルを元の位置にお戻しください. Return your seat and table to the original position during takeoff and landing." } ], "start": 201.79999999999998, "end": 224.0, "engine": "mlx-qwen3-asr", "language": null }, "cross_engine_agreement": 1.0, "replacement_language": "en", "replacement": "Return your seat and table to the original position during takeoff and landing.", "decision": "REPAIRABLE" }, "review_status": "CERTAIN_RESTORATION", "classification": "CERTAIN_RESTORATION", "raw_word_timestamps": [ { "word": "お", "start": 217.17999999999998, "end": 217.57999999999998, "probability": 0.12432861328125 }, { "word": "客", "start": 217.57999999999998, "end": 217.7, "probability": 0.11279296875 }, { "word": "様", "start": 217.7, "end": 218.14, "probability": 0.7119140625 }, { "word": "の", "start": 218.14, "end": 218.14, "probability": 0.39697265625 }, { "word": "お", "start": 218.14, "end": 218.23999999999998, "probability": 0.186767578125 }, { "word": "客", "start": 218.23999999999998, "end": 218.26, "probability": 0.7294921875 }, { "word": "様", "start": 218.26, "end": 218.88, "probability": 0.953125 }, { "word": "に", "start": 218.88, "end": 219.04, "probability": 0.29931640625 }, { "word": "お", "start": 219.04, "end": 219.04, "probability": 0.2135009765625 }, { "word": "客", "start": 219.04, "end": 219.55999999999997, "probability": 0.3974609375 }, { "word": "様", "start": 219.55999999999997, "end": 219.73999999999998, "probability": 0.98876953125 }, { "word": "に", "start": 219.73999999999998, "end": 219.76, "probability": 0.376953125 }, { "word": "お", "start": 219.76, "end": 219.76, "probability": 0.5107421875 }, { "word": "客", "start": 219.76, "end": 219.76, "probability": 0.62255859375 }, { "word": "様", "start": 219.76, "end": 220.27999999999997, "probability": 0.9951171875 }, { "word": "に", "start": 220.27999999999997, "end": 220.33999999999997, "probability": 0.50244140625 }, { "word": "お", "start": 220.33999999999997, "end": 220.51999999999998, "probability": 0.716796875 }, { "word": "客", "start": 220.51999999999998, "end": 220.55999999999997, "probability": 0.85986328125 }, { "word": "様", "start": 220.55999999999997, "end": 220.67999999999998, "probability": 0.99658203125 }, { "word": "に", "start": 220.67999999999998, "end": 220.67999999999998, "probability": 0.6103515625 }, { "word": "お", "start": 220.67999999999998, "end": 220.77999999999997, "probability": 0.8369140625 }, { "word": "客", "start": 220.77999999999997, "end": 220.77999999999997, "probability": 0.94873046875 }, { "word": "様", "start": 220.77999999999997, "end": 221.29999999999998, "probability": 0.99755859375 }, { "word": "に", "start": 221.29999999999998, "end": 221.29999999999998, "probability": 0.71142578125 }, { "word": "お", "start": 221.29999999999998, "end": 221.29999999999998, "probability": 0.90380859375 }, { "word": "客", "start": 221.29999999999998, "end": 221.29999999999998, "probability": 0.97509765625 }, { "word": "様", "start": 221.29999999999998, "end": 221.33999999999997, "probability": 0.998046875 }, { "word": "に", "start": 221.33999999999997, "end": 221.33999999999997, "probability": 0.7802734375 }, { "word": "お", "start": 221.33999999999997, "end": 221.33999999999997, "probability": 0.93798828125 }, { "word": "客", "start": 221.33999999999997, "end": 221.54, "probability": 0.9873046875 }, { "word": "様", "start": 221.54, "end": 221.54, "probability": 0.998046875 }, { "word": "に", "start": 221.54, "end": 221.54, "probability": 0.8251953125 }, { "word": "お", "start": 221.54, "end": 221.54, "probability": 0.95556640625 }, { "word": "客", "start": 221.54, "end": 221.61999999999998, "probability": 0.99169921875 }, { "word": "様", "start": 221.61999999999998, "end": 221.61999999999998, "probability": 0.99853515625 }, { "word": "に", "start": 221.61999999999998, "end": 221.61999999999998, "probability": 0.85302734375 }, { "word": "お", "start": 221.61999999999998, "end": 221.61999999999998, "probability": 0.96923828125 }, { "word": "客", "start": 221.61999999999998, "end": 221.64, "probability": 0.994140625 }, { "word": "様", "start": 221.64, "end": 221.82, "probability": 0.99853515625 }, { "word": "に", "start": 221.82, "end": 221.82, "probability": 0.8671875 }, { "word": "お", "start": 221.82, "end": 221.82, "probability": 0.974609375 }, { "word": "客", "start": 221.82, "end": 221.82, "probability": 0.99462890625 }, { "word": "様", "start": 221.82, "end": 221.82, "probability": 0.99853515625 }, { "word": "に", "start": 221.82, "end": 221.82, "probability": 0.8857421875 }, { "word": "お", "start": 221.82, "end": 222.04, "probability": 0.97705078125 }, { "word": "客", "start": 222.04, "end": 222.04, "probability": 0.99560546875 }, { "word": "様", "start": 222.04, "end": 222.04, "probability": 0.99853515625 }, { "word": "に", "start": 222.04, "end": 222.04, "probability": 0.89111328125 }, { "word": "お", "start": 222.04, "end": 222.04, "probability": 0.97998046875 }, { "word": "客", "start": 222.04, "end": 222.04, "probability": 0.99658203125 }, { "word": "様", "start": 222.04, "end": 222.04, "probability": 0.99853515625 }, { "word": "に", "start": 222.04, "end": 222.04, "probability": 0.900390625 }, { "word": "お", "start": 222.04, "end": 222.89999999999998, "probability": 0.98193359375 }, { "word": "客", "start": 222.89999999999998, "end": 222.89999999999998, "probability": 0.99658203125 }, { "word": "様", "start": 222.89999999999998, "end": 223.83999999999997, "probability": 0.99853515625 }, { "word": "に", "start": 223.83999999999997, "end": 223.83999999999997, "probability": 0.90283203125 }, { "word": "お", "start": 223.83999999999997, "end": 223.83999999999997, "probability": 0.98291015625 }, { "word": "客", "start": 223.83999999999997, "end": 223.83999999999997, "probability": 0.99658203125 }, { "word": "様", "start": 223.83999999999997, "end": 223.83999999999997, "probability": 0.99853515625 }, { "word": "に", "start": 223.83999999999997, "end": 223.83999999999997, "probability": 0.90771484375 }, { "word": "お", "start": 223.83999999999997, "end": 223.83999999999997, "probability": 0.98486328125 }, { "word": "客", "start": 223.83999999999997, "end": 223.83999999999997, "probability": 0.9970703125 }, { "word": "様", "start": 223.83999999999997, "end": 223.83999999999997, "probability": 0.99853515625 }, { "word": "に", "start": 223.83999999999997, "end": 223.83999999999997, "probability": 0.91259765625 }, { "word": "お", "start": 223.83999999999997, "end": 223.83999999999997, "probability": 0.98486328125 }, { "word": "客", "start": 223.83999999999997, "end": 223.83999999999997, "probability": 0.9970703125 }, { "word": "様", "start": 223.83999999999997, "end": 223.83999999999997, "probability": 0.99853515625 }, { "word": "に", "start": 223.83999999999997, "end": 223.83999999999997, "probability": 0.91748046875 }, { "word": "お", "start": 223.83999999999997, "end": 223.83999999999997, "probability": 0.98583984375 }, { "word": "客", "start": 223.83999999999997, "end": 223.83999999999997, "probability": 0.99755859375 }, { "word": "様", "start": 223.83999999999997, "end": 223.83999999999997, "probability": 0.99853515625 }, { "word": "に", "start": 223.83999999999997, "end": 223.83999999999997, "probability": 0.91943359375 }, { "word": "お", "start": 223.83999999999997, "end": 223.83999999999997, "probability": 0.986328125 }, { "word": "客", "start": 223.83999999999997, "end": 223.83999999999997, "probability": 0.9970703125 }, { "word": "様", "start": 223.83999999999997, "end": 223.83999999999997, "probability": 0.99853515625 }, { "word": "に", "start": 223.83999999999997, "end": 223.83999999999997, "probability": 0.91845703125 }, { "word": "お", "start": 223.83999999999997, "end": 223.83999999999997, "probability": 0.986328125 }, { "word": "客", "start": 223.83999999999997, "end": 223.83999999999997, "probability": 0.9970703125 }, { "word": "様", "start": 223.83999999999997, "end": 223.83999999999997, "probability": 0.99853515625 }, { "word": "に", "start": 223.83999999999997, "end": 223.83999999999997, "probability": 0.91552734375 }, { "word": "お", "start": 223.83999999999997, "end": 223.83999999999997, "probability": 0.9873046875 }, { "word": "客", "start": 223.83999999999997, "end": 223.83999999999997, "probability": 0.99755859375 }, { "word": "様", "start": 223.83999999999997, "end": 223.89999999999998, "probability": 0.9990234375 }, { "word": "に", "start": 223.89999999999998, "end": 223.89999999999998, "probability": 0.923828125 }, { "word": "お", "start": 223.89999999999998, "end": 223.98, "probability": 0.98681640625 }, { "word": "客", "start": 223.98, "end": 223.98, "probability": 0.9970703125 }, { "word": "様", "start": 223.98, "end": 223.98, "probability": 0.99853515625 }, { "word": "に", "start": 223.98, "end": 223.98, "probability": 0.919921875 }, { "word": "お", "start": 223.98, "end": 223.98, "probability": 0.98681640625 }, { "word": "客", "start": 223.98, "end": 223.98, "probability": 0.99755859375 }, { "word": "様", "start": 223.98, "end": 223.98, "probability": 0.9990234375 }, { "word": "に", "start": 223.98, "end": 223.98, "probability": 0.9228515625 }, { "word": "お", "start": 223.98, "end": 223.98, "probability": 0.98681640625 }, { "word": "客", "start": 223.98, "end": 223.98, "probability": 0.99755859375 }, { "word": "様", "start": 223.98, "end": 223.98, "probability": 0.9990234375 }, { "word": "に", "start": 223.98, "end": 223.98, "probability": 0.9228515625 }, { "word": "お", "start": 223.98, "end": 223.98, "probability": 0.9873046875 }, { "word": "客", "start": 223.98, "end": 223.98, "probability": 0.99755859375 }, { "word": "様", "start": 223.98, "end": 223.98, "probability": 0.9990234375 }, { "word": "に", "start": 223.98, "end": 223.98, "probability": 0.92138671875 }, { "word": "お", "start": 223.98, "end": 223.98, "probability": 0.9873046875 }, { "word": "客", "start": 223.98, "end": 223.98, "probability": 0.99755859375 }, { "word": "様", "start": 223.98, "end": 223.98, "probability": 0.9990234375 }, { "word": "に", "start": 223.98, "end": 223.98, "probability": 0.92431640625 }, { "word": "お", "start": 223.98, "end": 223.98, "probability": 0.98681640625 }, { "word": "客", "start": 223.98, "end": 223.98, "probability": 0.99755859375 }, { "word": "様", "start": 223.98, "end": 223.98, "probability": 0.99853515625 }, { "word": "に", "start": 223.98, "end": 223.98, "probability": 0.91845703125 }, { "word": "お", "start": 223.98, "end": 223.98, "probability": 0.986328125 }, { "word": "客", "start": 223.98, "end": 223.98, "probability": 0.99755859375 }, { "word": "様", "start": 223.98, "end": 223.98, "probability": 0.9990234375 }, { "word": "に", "start": 223.98, "end": 223.98, "probability": 0.916015625 }, { "word": "お", "start": 223.98, "end": 223.98, "probability": 0.986328125 }, { "word": "客", "start": 223.98, "end": 223.98, "probability": 0.99755859375 }, { "word": "様", "start": 223.98, "end": 223.98, "probability": 0.99853515625 }, { "word": "に", "start": 223.98, "end": 223.98, "probability": 0.9169921875 }, { "word": "お", "start": 223.98, "end": 223.98, "probability": 0.986328125 }, { "word": "客", "start": 223.98, "end": 223.98, "probability": 0.998046875 }, { "word": "様", "start": 223.98, "end": 223.98, "probability": 0.99853515625 }, { "word": "に", "start": 223.98, "end": 223.98, "probability": 0.9150390625 }, { "word": "お", "start": 223.98, "end": 223.98, "probability": 0.986328125 }, { "word": "客", "start": 223.98, "end": 223.98, "probability": 0.99755859375 }, { "word": "様", "start": 223.98, "end": 223.98, "probability": 0.99853515625 }, { "word": "に", "start": 223.98, "end": 223.98, "probability": 0.91796875 }, { "word": "お", "start": 223.98, "end": 223.98, "probability": 0.98583984375 }, { "word": "客", "start": 223.98, "end": 223.98, "probability": 0.99755859375 }, { "word": "様", "start": 223.98, "end": 223.98, "probability": 0.99853515625 }, { "word": "に", "start": 223.98, "end": 223.98, "probability": 0.91357421875 }, { "word": "お", "start": 223.98, "end": 223.98, "probability": 0.98583984375 }, { "word": "客", "start": 223.98, "end": 223.98, "probability": 0.99755859375 }, { "word": "様", "start": 223.98, "end": 223.98, "probability": 0.99853515625 }, { "word": "に", "start": 223.98, "end": 223.98, "probability": 0.91259765625 }, { "word": "お", "start": 223.98, "end": 223.98, "probability": 0.98583984375 }, { "word": "客", "start": 223.98, "end": 223.98, "probability": 0.99755859375 }, { "word": "様", "start": 223.98, "end": 223.98, "probability": 0.99853515625 }, { "word": "に", "start": 223.98, "end": 223.98, "probability": 0.91064453125 }, { "word": "お", "start": 223.98, "end": 223.98, "probability": 0.986328125 }, { "word": "客", "start": 223.98, "end": 223.98, "probability": 0.998046875 }, { "word": "様", "start": 223.98, "end": 223.98, "probability": 0.99853515625 }, { "word": "に", "start": 223.98, "end": 223.98, "probability": 0.9091796875 }, { "word": "お", "start": 223.98, "end": 223.98, "probability": 0.98583984375 }, { "word": "客", "start": 223.98, "end": 223.98, "probability": 0.99755859375 }, { "word": "様", "start": 223.98, "end": 223.98, "probability": 0.99853515625 }, { "word": "に", "start": 223.98, "end": 223.98, "probability": 0.90771484375 }, { "word": "お", "start": 223.98, "end": 223.98, "probability": 0.9853515625 }, { "word": "客", "start": 223.98, "end": 223.98, "probability": 0.99755859375 }, { "word": "様", "start": 223.98, "end": 223.98, "probability": 0.99853515625 }, { "word": "に", "start": 223.98, "end": 223.98, "probability": 0.90625 }, { "word": "お", "start": 223.98, "end": 223.98, "probability": 0.9853515625 }, { "word": "客", "start": 223.98, "end": 223.98, "probability": 0.99755859375 }, { "word": "様", "start": 223.98, "end": 223.98, "probability": 0.99853515625 }, { "word": "に", "start": 223.98, "end": 223.98, "probability": 0.9111328125 }, { "word": "お", "start": 223.98, "end": 223.98, "probability": 0.98486328125 }, { "word": "客", "start": 223.98, "end": 223.98, "probability": 0.99755859375 }, { "word": "様", "start": 223.98, "end": 223.98, "probability": 0.99853515625 }, { "word": "に", "start": 223.98, "end": 223.98, "probability": 0.91015625 }, { "word": "お", "start": 223.98, "end": 223.98, "probability": 0.9853515625 }, { "word": "客", "start": 223.98, "end": 223.98, "probability": 0.99755859375 }, { "word": "様", "start": 223.98, "end": 223.98, "probability": 0.99853515625 }, { "word": "に", "start": 223.98, "end": 223.98, "probability": 0.90771484375 }, { "word": "お", "start": 223.98, "end": 223.98, "probability": 0.98486328125 }, { "word": "客", "start": 223.98, "end": 223.98, "probability": 0.998046875 }, { "word": "様", "start": 223.98, "end": 223.98, "probability": 0.99853515625 }, { "word": "に", "start": 223.98, "end": 223.98, "probability": 0.90576171875 }, { "word": "お", "start": 223.98, "end": 223.98, "probability": 0.984375 }, { "word": "客", "start": 223.98, "end": 223.98, "probability": 0.99755859375 }, { "word": "様", "start": 223.98, "end": 223.98, "probability": 0.99853515625 }, { "word": "に", "start": 223.98, "end": 223.98, "probability": 0.908203125 }, { "word": "お", "start": 223.98, "end": 223.98, "probability": 0.98486328125 }, { "word": "客", "start": 223.98, "end": 223.98, "probability": 0.99755859375 }, { "word": "様", "start": 223.98, "end": 223.98, "probability": 0.99853515625 }, { "word": "に", "start": 223.98, "end": 223.98, "probability": 0.90234375 }, { "word": "お", "start": 223.98, "end": 223.98, "probability": 0.984375 }, { "word": "客", "start": 223.98, "end": 223.98, "probability": 0.99755859375 }, { "word": "様", "start": 223.98, "end": 223.98, "probability": 0.99853515625 }, { "word": "に", "start": 223.98, "end": 223.98, "probability": 0.904296875 }, { "word": "お", "start": 223.98, "end": 223.98, "probability": 0.9833984375 }, { "word": "客", "start": 223.98, "end": 223.98, "probability": 0.99755859375 }, { "word": "様", "start": 223.98, "end": 223.98, "probability": 0.99853515625 }, { "word": "に", "start": 223.98, "end": 223.98, "probability": 0.900390625 }, { "word": "お", "start": 223.98, "end": 223.98, "probability": 0.9833984375 }, { "word": "客", "start": 223.98, "end": 223.98, "probability": 0.99755859375 }, { "word": "様", "start": 223.98, "end": 223.98, "probability": 0.99853515625 }, { "word": "に", "start": 223.98, "end": 223.98, "probability": 0.90380859375 }, { "word": "お", "start": 223.98, "end": 223.98, "probability": 0.9833984375 }, { "word": "客", "start": 223.98, "end": 223.98, "probability": 0.99755859375 }, { "word": "様", "start": 223.98, "end": 223.98, "probability": 0.99853515625 }, { "word": "に", "start": 223.98, "end": 223.98, "probability": 0.90185546875 }, { "word": "お", "start": 223.98, "end": 223.98, "probability": 0.9833984375 }, { "word": "客", "start": 223.98, "end": 223.98, "probability": 0.99755859375 }, { "word": "様", "start": 223.98, "end": 223.98, "probability": 0.99853515625 }, { "word": "に", "start": 223.98, "end": 223.98, "probability": 0.908203125 }, { "word": "お", "start": 223.98, "end": 223.98, "probability": 0.98388671875 }, { "word": "客", "start": 223.98, "end": 223.98, "probability": 0.99755859375 }, { "word": "様", "start": 223.98, "end": 223.98, "probability": 0.99853515625 }, { "word": "に", "start": 223.98, "end": 223.98, "probability": 0.904296875 }, { "word": "お", "start": 223.98, "end": 223.98, "probability": 0.98388671875 }, { "word": "客", "start": 223.98, "end": 223.98, "probability": 0.99755859375 }, { "word": "様", "start": 223.98, "end": 223.98, "probability": 0.99853515625 }, { "word": "に", "start": 223.98, "end": 223.98, "probability": 0.9013671875 }, { "word": "お", "start": 223.98, "end": 223.98, "probability": 0.98388671875 }, { "word": "客", "start": 223.98, "end": 223.98, "probability": 0.9970703125 }, { "word": "様", "start": 223.98, "end": 223.98, "probability": 0.99853515625 }, { "word": "に", "start": 223.98, "end": 223.98, "probability": 0.8984375 }, { "word": "お", "start": 223.98, "end": 223.98, "probability": 0.9833984375 }, { "word": "客", "start": 223.98, "end": 223.98, "probability": 0.99755859375 }, { "word": "様", "start": 223.98, "end": 223.98, "probability": 0.99853515625 }, { "word": "に", "start": 223.98, "end": 223.98, "probability": 0.8974609375 }, { "word": "お", "start": 223.98, "end": 223.98, "probability": 0.98291015625 }, { "word": "客", "start": 223.98, "end": 223.98, "probability": 0.99755859375 }, { "word": "様", "start": 223.98, "end": 223.98, "probability": 0.99853515625 }, { "word": "に", "start": 223.98, "end": 223.98, "probability": 0.89794921875 }, { "word": "お", "start": 223.98, "end": 223.98, "probability": 0.9833984375 }, { "word": "客", "start": 223.98, "end": 223.98, "probability": 0.99755859375 }, { "word": "様", "start": 223.98, "end": 223.98, "probability": 0.99853515625 }, { "word": "に", "start": 223.98, "end": 223.98, "probability": 0.89990234375 }, { "word": "お", "start": 223.98, "end": 223.98, "probability": 0.98291015625 }, { "word": "客", "start": 223.98, "end": 223.98, "probability": 0.99755859375 }, { "word": "様", "start": 223.98, "end": 223.98, "probability": 0.99853515625 }, { "word": "に", "start": 223.98, "end": 223.98, "probability": 0.89599609375 }, { "word": "お", "start": 223.98, "end": 223.98, "probability": 0.98291015625 }, { "word": "客", "start": 223.98, "end": 223.98, "probability": 0.9970703125 }, { "word": "様", "start": 223.98, "end": 223.98, "probability": 0.99853515625 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "packet_ids": [ 12 ], "index": 38, "source_text": "Return your seat and table to the original position during takeoff and landing.", "confidence": 0.98, "correction_reason": "Independent ASR confirms exact wording despite raw ASR failure." }, { "start": 223.8, "end": 229.34, "text": "ベルト着用サイン点灯中並びにお座りの間は常にシートベルトをお締めください", "raw_asr_text": "ベルト着用サイン点灯中並びにお座りの間は常にシートベルトをお締めください", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 13, "words": [], "asr_metrics": { "avg_logprob": -0.3144910729449728, "compression_ratio": 1.4377880184331797, "no_speech_prob": 6.32960003854599e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "ベルト着用サイン点灯中並びにお座りの間は常にシートベルトをお締めください外す場合は金具の片方を持ち上げてくださいこの便は全席禁煙です化粧室でのタバコは航空法で禁止されています携帯のタバコはステーキのタバコですゲータイム" }, { "language": "en", "text": "The seat belt is closed. Please keep your seat belt fast whenever you are seated. The seat belt is closed. The seat belt is closed. The bathroom is closed. The bathroom is prohibited in the cabin including the lavatory. Okay. Okay." } ], "id": 39, "review_annotation": "", "row_id": 39, "bounded_row_id": 39, "parent_row_ids": [ 39 ], "parent_row_id": 39, "packet_id": 13, "LID_evidence": [ { "region_id": 32, "probabilities": { "ja": 0.1611263426080661, "zh": 0.0040083565956378795, "en": 0.3140458025753854 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 32 ], "language_confidence": 0.1611263426080661, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 223.0, "packet_end": 249.60000000000002, "raw_text": "ベルト着用サイン点灯中並びにお座りの間は常にシートベルトをお締めください", "selected_text": "ベルト着用サイン点灯中並びにお座りの間は常にシートベルトをお締めください", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 39 ], "packet_id": 13, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "ベルト着用サイン点灯中並びにお座りの間は常にシートベルトをお締めください", "after": "ベルト着用サイン点灯中並びにお座りの間は常にシートベルトをお締めください", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard safety announcement phrasing matches context." } ], "language": "ja", "packet_ids": [ 13 ], "index": 39, "source_text": "ベルト着用サイン点灯中並びにお座りの間は常にシートベルトをお締めください", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Standard safety announcement phrasing matches context.", "raw_word_timestamps": [ { "word": "ベ", "start": 223.68, "end": 224.0, "probability": 0.8447265625 }, { "word": "ル", "start": 224.0, "end": 224.1, "probability": 0.99951171875 }, { "word": "ト", "start": 224.1, "end": 224.2, "probability": 0.99951171875 }, { "word": "着", "start": 224.2, "end": 224.34, "probability": 0.99462890625 }, { "word": "用", "start": 224.34, "end": 224.6, "probability": 0.99951171875 }, { "word": "サ", "start": 224.6, "end": 224.8, "probability": 0.9990234375 }, { "word": "イ", "start": 224.8, "end": 224.9, "probability": 1.0 }, { "word": "ン", "start": 224.9, "end": 224.96, "probability": 0.9990234375 }, { "word": "点", "start": 224.96, "end": 225.12, "probability": 0.91259765625 }, { "word": "灯", "start": 225.12, "end": 225.38, "probability": 0.99560546875 }, { "word": "中", "start": 225.38, "end": 225.64, "probability": 0.998046875 }, { "word": "並", "start": 225.64, "end": 226.14, "probability": 0.86865234375 }, { "word": "び", "start": 226.14, "end": 226.28, "probability": 0.99853515625 }, { "word": "に", "start": 226.28, "end": 226.42, "probability": 0.99951171875 }, { "word": "お", "start": 226.42, "end": 226.6, "probability": 0.99658203125 }, { "word": "座", "start": 226.6, "end": 226.78, "probability": 0.9931640625 }, { "word": "り", "start": 226.78, "end": 226.9, "probability": 1.0 }, { "word": "の", "start": 226.9, "end": 227.0, "probability": 0.99951171875 }, { "word": "間", "start": 227.0, "end": 227.16, "probability": 1.0 }, { "word": "は", "start": 227.16, "end": 227.44, "probability": 0.99951171875 }, { "word": "常", "start": 227.44, "end": 227.8, "probability": 0.97802734375 }, { "word": "に", "start": 227.8, "end": 227.98, "probability": 1.0 }, { "word": "シ", "start": 227.98, "end": 228.14, "probability": 0.998046875 }, { "word": "ート", "start": 228.14, "end": 228.28, "probability": 1.0 }, { "word": "ベ", "start": 228.28, "end": 228.44, "probability": 0.99951171875 }, { "word": "ル", "start": 228.44, "end": 228.5, "probability": 1.0 }, { "word": "ト", "start": 228.5, "end": 228.62, "probability": 1.0 }, { "word": "を", "start": 228.62, "end": 228.74, "probability": 1.0 }, { "word": "お", "start": 228.74, "end": 228.86, "probability": 0.98828125 }, { "word": "締", "start": 228.86, "end": 228.9, "probability": 0.9716796875 }, { "word": "め", "start": 228.9, "end": 229.0, "probability": 1.0 }, { "word": "ください", "start": 229.0, "end": 229.34, "probability": 0.99365234375 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 229.34, "end": 236.1, "text": "外す場合は金具の片方を持ち上げてください", "raw_asr_text": "外す場合は金具の片方を持ち上げてください", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 13, "words": [ { "word": "外", "start": 229.34, "end": 229.8, "probability": 0.86328125 }, { "word": "す", "start": 229.8, "end": 233.5, "probability": 0.9990234375 }, { "word": "場", "start": 233.5, "end": 233.68, "probability": 0.99951171875 }, { "word": "合", "start": 233.68, "end": 233.82, "probability": 1.0 }, { "word": "は", "start": 233.82, "end": 234.02, "probability": 1.0 }, { "word": "金", "start": 234.02, "end": 234.38, "probability": 0.99658203125 }, { "word": "具", "start": 234.38, "end": 234.52, "probability": 1.0 }, { "word": "の", "start": 234.52, "end": 234.64, "probability": 1.0 }, { "word": "片", "start": 234.64, "end": 234.8, "probability": 0.99951171875 }, { "word": "方", "start": 234.8, "end": 235.02, "probability": 1.0 }, { "word": "を", "start": 235.02, "end": 235.18, "probability": 1.0 }, { "word": "持", "start": 235.18, "end": 235.32, "probability": 0.99951171875 }, { "word": "ち", "start": 235.32, "end": 235.42, "probability": 1.0 }, { "word": "上", "start": 235.42, "end": 235.5, "probability": 0.99951171875 }, { "word": "げ", "start": 235.5, "end": 235.6, "probability": 1.0 }, { "word": "て", "start": 235.6, "end": 235.7, "probability": 1.0 }, { "word": "ください", "start": 235.7, "end": 236.1, "probability": 0.99951171875 } ], "asr_metrics": { "avg_logprob": -0.3144910729449728, "compression_ratio": 1.4377880184331797, "no_speech_prob": 6.32960003854599e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "ベルト着用サイン点灯中並びにお座りの間は常にシートベルトをお締めください外す場合は金具の片方を持ち上げてくださいこの便は全席禁煙です化粧室でのタバコは航空法で禁止されています携帯のタバコはステーキのタバコですゲータイム" }, { "language": "en", "text": "The seat belt is closed. Please keep your seat belt fast whenever you are seated. The seat belt is closed. The seat belt is closed. The bathroom is closed. The bathroom is prohibited in the cabin including the lavatory. Okay. Okay." } ], "id": 40, "review_annotation": "", "row_id": 40, "bounded_row_id": 40, "parent_row_ids": [ 40 ], "parent_row_id": 40, "packet_id": 13, "LID_evidence": [ { "region_id": 32, "probabilities": { "ja": 0.1611263426080661, "zh": 0.0040083565956378795, "en": 0.3140458025753854 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 32 ], "language_confidence": 0.1611263426080661, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 223.0, "packet_end": 249.60000000000002, "raw_text": "外す場合は金具の片方を持ち上げてください", "selected_text": "外す場合は金具の片方を持ち上げてください", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 40 ], "packet_id": 13, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "外す場合は金具の片方を持ち上げてください", "after": "外す場合は金具の片方を持ち上げてください", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard safety instruction phrasing matches context and ASR output perfectly." } ], "language": "ja", "packet_ids": [ 13 ], "index": 40, "source_text": "外す場合は金具の片方を持ち上げてください", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Standard safety instruction phrasing matches context and ASR output perfectly." }, { "start": 239.1, "end": 241.04, "text": "この便は全席禁煙です", "raw_asr_text": "この便は全席禁煙です", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 13, "words": [ { "word": "この", "start": 239.1, "end": 239.42000000000002, "probability": 0.9931640625 }, { "word": "便", "start": 239.42000000000002, "end": 239.57999999999998, "probability": 0.5908203125 }, { "word": "は", "start": 239.57999999999998, "end": 239.84, "probability": 1.0 }, { "word": "全", "start": 239.84, "end": 240.1, "probability": 0.94091796875 }, { "word": "席", "start": 240.1, "end": 240.38, "probability": 0.99609375 }, { "word": "禁", "start": 240.38, "end": 240.62, "probability": 0.9990234375 }, { "word": "煙", "start": 240.62, "end": 240.76, "probability": 0.9931640625 }, { "word": "です", "start": 240.76, "end": 241.04, "probability": 1.0 } ], "asr_metrics": { "avg_logprob": -0.3144910729449728, "compression_ratio": 1.4377880184331797, "no_speech_prob": 6.32960003854599e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "ベルト着用サイン点灯中並びにお座りの間は常にシートベルトをお締めください外す場合は金具の片方を持ち上げてくださいこの便は全席禁煙です化粧室でのタバコは航空法で禁止されています携帯のタバコはステーキのタバコですゲータイム" }, { "language": "en", "text": "The seat belt is closed. Please keep your seat belt fast whenever you are seated. The seat belt is closed. The seat belt is closed. The bathroom is closed. The bathroom is prohibited in the cabin including the lavatory. Okay. Okay." } ], "id": 41, "review_annotation": "", "row_id": 41, "bounded_row_id": 41, "parent_row_ids": [ 41 ], "parent_row_id": 41, "packet_id": 13, "LID_evidence": [ { "region_id": 33, "probabilities": { "ja": 0.23317185952328146, "zh": 0.0036296734961069888, "en": 0.07762659659783822 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 33 ], "language_confidence": 0.23317185952328146, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 223.0, "packet_end": 249.60000000000002, "raw_text": "この便は全席禁煙です", "selected_text": "この便は全席禁煙です", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 41 ], "packet_id": 13, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "この便は全席禁煙です", "after": "この便は全席禁煙です", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard announcement phrase, clearly supported by context and ASR." } ], "language": "ja", "packet_ids": [ 13 ], "index": 41, "source_text": "この便は全席禁煙です", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Standard announcement phrase, clearly supported by context and ASR." }, { "start": 241.7, "end": 244.88, "text": "化粧室でのタバコは航空法で禁止されています", "raw_asr_text": "化粧室でのタバコは航空法で禁止されています", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 13, "words": [ { "word": "化", "start": 241.7, "end": 242.02, "probability": 0.96923828125 }, { "word": "粧", "start": 242.02, "end": 242.22, "probability": 0.9990234375 }, { "word": "室", "start": 242.22, "end": 242.38, "probability": 0.9931640625 }, { "word": "で", "start": 242.38, "end": 242.5, "probability": 0.99951171875 }, { "word": "の", "start": 242.5, "end": 242.62, "probability": 1.0 }, { "word": "タ", "start": 242.62, "end": 242.74, "probability": 0.689453125 }, { "word": "バ", "start": 242.74, "end": 242.82, "probability": 0.9990234375 }, { "word": "コ", "start": 242.82, "end": 242.92000000000002, "probability": 0.9990234375 }, { "word": "は", "start": 242.92000000000002, "end": 243.1, "probability": 1.0 }, { "word": "航", "start": 243.1, "end": 243.42000000000002, "probability": 0.9921875 }, { "word": "空", "start": 243.42000000000002, "end": 243.57999999999998, "probability": 0.99951171875 }, { "word": "法", "start": 243.57999999999998, "end": 243.74, "probability": 0.82470703125 }, { "word": "で", "start": 243.74, "end": 243.9, "probability": 1.0 }, { "word": "禁", "start": 243.9, "end": 244.12, "probability": 1.0 }, { "word": "止", "start": 244.12, "end": 244.22, "probability": 1.0 }, { "word": "さ", "start": 244.22, "end": 244.36, "probability": 1.0 }, { "word": "れて", "start": 244.36, "end": 244.5, "probability": 1.0 }, { "word": "います", "start": 244.5, "end": 244.88, "probability": 1.0 } ], "asr_metrics": { "avg_logprob": -0.3144910729449728, "compression_ratio": 1.4377880184331797, "no_speech_prob": 6.32960003854599e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "ベルト着用サイン点灯中並びにお座りの間は常にシートベルトをお締めください外す場合は金具の片方を持ち上げてくださいこの便は全席禁煙です化粧室でのタバコは航空法で禁止されています携帯のタバコはステーキのタバコですゲータイム" }, { "language": "en", "text": "The seat belt is closed. Please keep your seat belt fast whenever you are seated. The seat belt is closed. The seat belt is closed. The bathroom is closed. The bathroom is prohibited in the cabin including the lavatory. Okay. Okay." } ], "id": 42, "review_annotation": "", "row_id": 42, "bounded_row_id": 42, "parent_row_ids": [ 42 ], "parent_row_id": 42, "packet_id": 13, "LID_evidence": [ { "region_id": 34, "probabilities": { "ja": 0.09312978453817777, "zh": 0.002817079406263979, "en": 0.3561339686986887 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 34 ], "language_confidence": 0.09312978453817777, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 223.0, "packet_end": 249.60000000000002, "raw_text": "化粧室でのタバコは航空法で禁止されています", "selected_text": "化粧室でのタバコは航空法で禁止されています", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 42 ], "packet_id": 13, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "化粧室でのタバコは航空法で禁止されています", "after": "化粧室でのタバコは航空法で禁止されています", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard legal warning phrase, consistent with ASR and context." } ], "language": "ja", "packet_ids": [ 13 ], "index": 42, "source_text": "化粧室でのタバコは航空法で禁止されています", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Standard legal warning phrase, consistent with ASR and context." }, { "start": 245.22, "end": 248.5, "text": "携帯のタバコはスモーキングです", "raw_asr_text": "携帯のタバコはステーキのタバコです", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 13, "words": [], "asr_metrics": { "avg_logprob": -0.3144910729449728, "compression_ratio": 1.4377880184331797, "no_speech_prob": 6.32960003854599e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "ベルト着用サイン点灯中並びにお座りの間は常にシートベルトをお締めください外す場合は金具の片方を持ち上げてくださいこの便は全席禁煙です化粧室でのタバコは航空法で禁止されています携帯のタバコはステーキのタバコですゲータイム" }, { "language": "en", "text": "The seat belt is closed. Please keep your seat belt fast whenever you are seated. The seat belt is closed. The seat belt is closed. The bathroom is closed. The bathroom is prohibited in the cabin including the lavatory. Okay. Okay." } ], "id": 43, "review_annotation": "待確認", "row_id": 43, "bounded_row_id": 43, "parent_row_ids": [ 43 ], "parent_row_id": 43, "packet_id": 13, "LID_evidence": [ { "region_id": 34, "probabilities": { "ja": 0.09312978453817777, "zh": 0.002817079406263979, "en": 0.3561339686986887 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 34 ], "language_confidence": 0.09312978453817777, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 223.0, "packet_end": 249.60000000000002, "raw_text": "携帯のタバコはステーキのタバコです", "selected_text": "携帯のタバコはスモーキングです", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 43 ], "packet_id": 13, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "携帯のタバコはステーキのタバコです", "after": "携帯のタバコはスモーキングです", "classification": "BEST_SUPPORTED_WITH_REVIEW", "reason": "Independent ASR 'スモーキング' and context '禁煙' support 'smoking' over 'steak'." } ], "language": "ja", "packet_ids": [ 13 ], "index": 43, "independent_ASR": { "engine": "kotoba-whisper", "model_path": "/Users/sagawa/AI_WORK/cache/huggingface/hub/models--kaiinui--kotoba-whisper-v2.0-mlx/snapshots/1a5ac6e52394898ef8996212e849d964e0dfb9ae", "language": "ja", "audio_interval": [ 223.0, 249.60000000000002 ], "target_interval": [ 223.8, 248.8 ], "text": "ト着用サイン点灯中並びにお座りの間は常にシートベルトをお締めください外す場合は金具の片方を持ち上げてください携帯は全席禁煙です!化粧室でのタバコは航空法で禁止されていますスモーキングがコーヒベントリー", "words": [ { "word": "ト", "start": 224.14, "end": 224.24, "probability": 0.99853515625 }, { "word": "着", "start": 224.24, "end": 224.38, "probability": 0.9990234375 }, { "word": "用", "start": 224.38, "end": 224.64, "probability": 1.0 }, { "word": "サ", "start": 224.64, "end": 224.92, "probability": 0.99755859375 }, { "word": "イ", "start": 224.92, "end": 224.92, "probability": 1.0 }, { "word": "ン", "start": 224.92, "end": 225.0, "probability": 1.0 }, { "word": "点", "start": 225.0, "end": 225.3, "probability": 0.70654296875 }, { "word": "灯", "start": 225.3, "end": 225.32, "probability": 0.98681640625 }, { "word": "中", "start": 225.32, "end": 225.7, "probability": 0.9990234375 }, { "word": "並", "start": 225.7, "end": 226.2, "probability": 0.42431640625 }, { "word": "び", "start": 226.2, "end": 226.32, "probability": 0.99853515625 }, { "word": "に", "start": 226.32, "end": 226.46, "probability": 0.99951171875 }, { "word": "お", "start": 226.46, "end": 226.66, "probability": 0.998046875 }, { "word": "座", "start": 226.66, "end": 226.78, "probability": 0.9970703125 }, { "word": "り", "start": 226.78, "end": 226.92, "probability": 1.0 }, { "word": "の", "start": 226.92, "end": 227.02, "probability": 0.99951171875 }, { "word": "間", "start": 227.02, "end": 227.22, "probability": 0.9990234375 }, { "word": "は", "start": 227.22, "end": 227.62, "probability": 1.0 }, { "word": "常", "start": 227.62, "end": 227.86, "probability": 0.7470703125 }, { "word": "に", "start": 227.86, "end": 228.0, "probability": 1.0 }, { "word": "シ", "start": 228.0, "end": 228.18, "probability": 0.9951171875 }, { "word": "ート", "start": 228.18, "end": 228.3, "probability": 1.0 }, { "word": "ベ", "start": 228.3, "end": 228.52, "probability": 0.99951171875 }, { "word": "ル", "start": 228.52, "end": 228.56, "probability": 1.0 }, { "word": "ト", "start": 228.56, "end": 228.68, "probability": 0.99951171875 }, { "word": "を", "start": 228.68, "end": 228.76, "probability": 0.99951171875 }, { "word": "お", "start": 228.76, "end": 228.88, "probability": 0.83935546875 }, { "word": "締", "start": 228.88, "end": 228.94, "probability": 0.8564453125 }, { "word": "め", "start": 228.94, "end": 229.04, "probability": 1.0 }, { "word": "ください", "start": 229.04, "end": 229.46, "probability": 0.99365234375 }, { "word": "外", "start": 229.46, "end": 229.84, "probability": 0.43798828125 }, { "word": "す", "start": 229.84, "end": 233.5, "probability": 1.0 }, { "word": "場", "start": 233.5, "end": 233.72, "probability": 0.99951171875 }, { "word": "合", "start": 233.72, "end": 234.06, "probability": 1.0 }, { "word": "は", "start": 234.06, "end": 234.22, "probability": 1.0 }, { "word": "金", "start": 234.22, "end": 234.44, "probability": 0.9541015625 }, { "word": "具", "start": 234.44, "end": 234.54, "probability": 0.99560546875 }, { "word": "の", "start": 234.54, "end": 234.68, "probability": 1.0 }, { "word": "片", "start": 234.68, "end": 234.88, "probability": 0.99951171875 }, { "word": "方", "start": 234.88, "end": 235.1, "probability": 1.0 }, { "word": "を", "start": 235.1, "end": 235.2, "probability": 0.9990234375 }, { "word": "持", "start": 235.2, "end": 235.4, "probability": 0.99853515625 }, { "word": "ち", "start": 235.4, "end": 235.46, "probability": 1.0 }, { "word": "上", "start": 235.46, "end": 235.56, "probability": 1.0 }, { "word": "げ", "start": 235.56, "end": 235.66, "probability": 1.0 }, { "word": "て", "start": 235.66, "end": 235.76, "probability": 1.0 }, { "word": "ください", "start": 235.76, "end": 236.16, "probability": 0.9990234375 }, { "word": "携", "start": 238.88, "end": 239.24, "probability": 0.58935546875 }, { "word": "帯", "start": 239.24, "end": 239.56, "probability": 0.9990234375 }, { "word": "は", "start": 239.56, "end": 239.86, "probability": 0.282470703125 }, { "word": "全", "start": 239.86, "end": 240.12, "probability": 0.958984375 }, { "word": "席", "start": 240.12, "end": 240.36, "probability": 0.62255859375 }, { "word": "禁", "start": 240.36, "end": 240.62, "probability": 0.998046875 }, { "word": "煙", "start": 240.62, "end": 240.74, "probability": 0.99609375 }, { "word": "です!", "start": 240.74, "end": 241.02, "probability": 0.99951171875 }, { "word": "化", "start": 242.02, "end": 242.02, "probability": 0.00490570068359375 }, { "word": "粧", "start": 242.02, "end": 242.22, "probability": 0.998046875 }, { "word": "室", "start": 242.22, "end": 242.4, "probability": 0.9990234375 }, { "word": "で", "start": 242.4, "end": 242.5, "probability": 0.998046875 }, { "word": "の", "start": 242.5, "end": 242.62, "probability": 1.0 }, { "word": "タ", "start": 242.62, "end": 242.76, "probability": 0.958984375 }, { "word": "バ", "start": 242.76, "end": 242.84, "probability": 0.99951171875 }, { "word": "コ", "start": 242.84, "end": 242.92000000000002, "probability": 1.0 }, { "word": "は", "start": 242.92000000000002, "end": 243.1, "probability": 0.99951171875 }, { "word": "航", "start": 243.1, "end": 243.46, "probability": 0.9990234375 }, { "word": "空", "start": 243.46, "end": 243.6, "probability": 1.0 }, { "word": "法", "start": 243.6, "end": 243.78, "probability": 0.5556640625 }, { "word": "で", "start": 243.78, "end": 243.94, "probability": 0.99951171875 }, { "word": "禁", "start": 243.94, "end": 244.16, "probability": 1.0 }, { "word": "止", "start": 244.16, "end": 244.22, "probability": 1.0 }, { "word": "さ", "start": 244.22, "end": 244.38, "probability": 1.0 }, { "word": "れて", "start": 244.38, "end": 244.52, "probability": 1.0 }, { "word": "います", "start": 244.52, "end": 244.9, "probability": 0.99951171875 }, { "word": "ス", "start": 245.18, "end": 245.42000000000002, "probability": 0.1824951171875 }, { "word": "モ", "start": 245.42000000000002, "end": 245.52, "probability": 0.9423828125 }, { "word": "ー", "start": 245.52, "end": 245.66, "probability": 0.890625 }, { "word": "キ", "start": 245.66, "end": 245.68, "probability": 0.990234375 }, { "word": "ング", "start": 245.68, "end": 245.76, "probability": 0.9892578125 }, { "word": "が", "start": 245.76, "end": 245.88, "probability": 0.515625 }, { "word": " コ", "start": 246.72, "end": 246.84, "probability": 0.1697998046875 }, { "word": "ーヒ", "start": 246.84, "end": 247.0, "probability": 0.5673828125 }, { "word": "ベ", "start": 247.0, "end": 247.0, "probability": 0.1234130859375 }, { "word": "ント", "start": 247.0, "end": 247.1, "probability": 0.60986328125 }, { "word": "リ", "start": 247.1, "end": 247.1, "probability": 0.9990234375 }, { "word": "ー", "start": 247.1, "end": 248.5, "probability": 0.99951171875 } ], "raw_result": { "text": "ベルト着用サイン点灯中並びにお座りの間は常にシートベルトをお締めください外す場合は金具の片方を持ち上げてください携帯は全席禁煙です!化粧室でのタバコは航空法で禁止されていますスモーキングが コーヒベントリー", "segments": [ { "id": 0, "seek": 0, "start": 0.0, "end": 2.7, "text": "ベルト着用サイン点灯中", "tokens": [ 50365, 45290, 9405, 7588, 20708, 9254, 23607, 8040, 4824, 12579, 13290, 107, 5975, 50499 ], "temperature": 0.0, "avg_logprob": NaN, "compression_ratio": 1.2885572139303483, "no_speech_prob": 0.09440445899963379, "words": [ { "word": "ベ", "start": 0.0, "end": 0.0, "probability": 0.92919921875 }, { "word": "ル", "start": 0.0, "end": 1.14, "probability": 0.99951171875 }, { "word": "ト", "start": 1.14, "end": 1.24, "probability": 0.99853515625 }, { "word": "着", "start": 1.24, "end": 1.38, "probability": 0.9990234375 }, { "word": "用", "start": 1.38, "end": 1.64, "probability": 1.0 }, { "word": "サ", "start": 1.64, "end": 1.92, "probability": 0.99755859375 }, { "word": "イ", "start": 1.92, "end": 1.92, "probability": 1.0 }, { "word": "ン", "start": 1.92, "end": 2.0, "probability": 1.0 }, { "word": "点", "start": 2.0, "end": 2.3, "probability": 0.70654296875 }, { "word": "灯", "start": 2.3, "end": 2.32, "probability": 0.98681640625 }, { "word": "中", "start": 2.32, "end": 2.7, "probability": 0.9990234375 } ] }, { "id": 1, "seek": 0, "start": 2.7, "end": 4.62, "text": "並びにお座りの間は", "tokens": [ 50499, 35934, 38761, 4108, 6117, 6346, 100, 5095, 2972, 11016, 3065, 50593 ], "temperature": 0.0, "avg_logprob": NaN, "compression_ratio": 1.2885572139303483, "no_speech_prob": 0.09440445899963379, "words": [ { "word": "並", "start": 2.7, "end": 3.2, "probability": 0.42431640625 }, { "word": "び", "start": 3.2, "end": 3.32, "probability": 0.99853515625 }, { "word": "に", "start": 3.32, "end": 3.46, "probability": 0.99951171875 }, { "word": "お", "start": 3.46, "end": 3.66, "probability": 0.998046875 }, { "word": "座", "start": 3.66, "end": 3.78, "probability": 0.9970703125 }, { "word": "り", "start": 3.78, "end": 3.92, "probability": 1.0 }, { "word": "の", "start": 3.92, "end": 4.02, "probability": 0.99951171875 }, { "word": "間", "start": 4.02, "end": 4.22, "probability": 0.9990234375 }, { "word": "は", "start": 4.22, "end": 4.62, "probability": 1.0 } ] }, { "id": 2, "seek": 0, "start": 4.62, "end": 6.46, "text": "常にシートベルトをお締めください", "tokens": [ 50593, 11279, 4108, 11054, 38551, 45290, 9405, 7588, 5998, 6117, 12608, 254, 11429, 25079, 50685 ], "temperature": 0.0, "avg_logprob": NaN, "compression_ratio": 1.2885572139303483, "no_speech_prob": 0.09440445899963379, "words": [ { "word": "常", "start": 4.62, "end": 4.86, "probability": 0.7470703125 }, { "word": "に", "start": 4.86, "end": 5.0, "probability": 1.0 }, { "word": "シ", "start": 5.0, "end": 5.18, "probability": 0.9951171875 }, { "word": "ート", "start": 5.18, "end": 5.3, "probability": 1.0 }, { "word": "ベ", "start": 5.3, "end": 5.52, "probability": 0.99951171875 }, { "word": "ル", "start": 5.52, "end": 5.56, "probability": 1.0 }, { "word": "ト", "start": 5.56, "end": 5.68, "probability": 0.99951171875 }, { "word": "を", "start": 5.68, "end": 5.76, "probability": 0.99951171875 }, { "word": "お", "start": 5.76, "end": 5.88, "probability": 0.83935546875 }, { "word": "締", "start": 5.88, "end": 5.94, "probability": 0.8564453125 }, { "word": "め", "start": 5.94, "end": 6.04, "probability": 1.0 }, { "word": "ください", "start": 6.04, "end": 6.46, "probability": 0.99365234375 } ] }, { "id": 3, "seek": 0, "start": 6.46, "end": 13.16, "text": "外す場合は金具の片方を持ち上げてください", "tokens": [ 50685, 12022, 2659, 21446, 14245, 3065, 19117, 39806, 2972, 16668, 9249, 5998, 17694, 6574, 5708, 26012, 2996, 25079, 51023 ], "temperature": 0.0, "avg_logprob": NaN, "compression_ratio": 1.2885572139303483, "no_speech_prob": 0.09440445899963379, "words": [ { "word": "外", "start": 6.46, "end": 6.84, "probability": 0.43798828125 }, { "word": "す", "start": 6.84, "end": 10.5, "probability": 1.0 }, { "word": "場", "start": 10.5, "end": 10.72, "probability": 0.99951171875 }, { "word": "合", "start": 10.72, "end": 11.06, "probability": 1.0 }, { "word": "は", "start": 11.06, "end": 11.22, "probability": 1.0 }, { "word": "金", "start": 11.22, "end": 11.44, "probability": 0.9541015625 }, { "word": "具", "start": 11.44, "end": 11.54, "probability": 0.99560546875 }, { "word": "の", "start": 11.54, "end": 11.68, "probability": 1.0 }, { "word": "片", "start": 11.68, "end": 11.88, "probability": 0.99951171875 }, { "word": "方", "start": 11.88, "end": 12.1, "probability": 1.0 }, { "word": "を", "start": 12.1, "end": 12.2, "probability": 0.9990234375 }, { "word": "持", "start": 12.2, "end": 12.4, "probability": 0.99853515625 }, { "word": "ち", "start": 12.4, "end": 12.46, "probability": 1.0 }, { "word": "上", "start": 12.46, "end": 12.56, "probability": 1.0 }, { "word": "げ", "start": 12.56, "end": 12.66, "probability": 1.0 }, { "word": "て", "start": 12.66, "end": 12.76, "probability": 1.0 }, { "word": "ください", "start": 12.76, "end": 13.16, "probability": 0.9990234375 } ] }, { "id": 4, "seek": 0, "start": 15.879999999999999, "end": 18.02, "text": "携帯は全席禁煙です", "tokens": [ 51023, 22220, 118, 4845, 107, 3065, 11319, 4845, 255, 18614, 223, 17867, 247, 4767, 51267 ], "temperature": 0.0, "avg_logprob": NaN, "compression_ratio": 1.2885572139303483, "no_speech_prob": 0.09440445899963379, "words": [ { "word": "携", "start": 15.879999999999999, "end": 16.24, "probability": 0.58935546875 }, { "word": "帯", "start": 16.24, "end": 16.56, "probability": 0.9990234375 }, { "word": "は", "start": 16.56, "end": 16.86, "probability": 0.282470703125 }, { "word": "全", "start": 16.86, "end": 17.12, "probability": 0.958984375 }, { "word": "席", "start": 17.12, "end": 17.36, "probability": 0.62255859375 }, { "word": "禁", "start": 17.36, "end": 17.62, "probability": 0.998046875 }, { "word": "煙", "start": 17.62, "end": 17.74, "probability": 0.99609375 }, { "word": "です!", "start": 17.74, "end": 18.02, "probability": 0.99951171875 } ] }, { "id": 5, "seek": 0, "start": 19.02, "end": 21.9, "text": "!化粧室でのタバコは航空法で禁止されています", "tokens": [ 51267, 0, 23756, 18393, 100, 2415, 97, 2474, 2972, 12144, 18593, 18066, 3065, 10256, 103, 23322, 11148, 2474, 18614, 223, 30438, 6722, 26638, 11267, 51459 ], "temperature": 0.0, "avg_logprob": NaN, "compression_ratio": 1.2885572139303483, "no_speech_prob": 0.09440445899963379, "words": [ { "word": "化", "start": 19.02, "end": 19.02, "probability": 0.00490570068359375 }, { "word": "粧", "start": 19.02, "end": 19.22, "probability": 0.998046875 }, { "word": "室", "start": 19.22, "end": 19.4, "probability": 0.9990234375 }, { "word": "で", "start": 19.4, "end": 19.5, "probability": 0.998046875 }, { "word": "の", "start": 19.5, "end": 19.62, "probability": 1.0 }, { "word": "タ", "start": 19.62, "end": 19.76, "probability": 0.958984375 }, { "word": "バ", "start": 19.76, "end": 19.84, "probability": 0.99951171875 }, { "word": "コ", "start": 19.84, "end": 19.92, "probability": 1.0 }, { "word": "は", "start": 19.92, "end": 20.1, "probability": 0.99951171875 }, { "word": "航", "start": 20.1, "end": 20.46, "probability": 0.9990234375 }, { "word": "空", "start": 20.46, "end": 20.6, "probability": 1.0 }, { "word": "法", "start": 20.6, "end": 20.78, "probability": 0.5556640625 }, { "word": "で", "start": 20.78, "end": 20.94, "probability": 0.99951171875 }, { "word": "禁", "start": 20.94, "end": 21.16, "probability": 1.0 }, { "word": "止", "start": 21.16, "end": 21.22, "probability": 1.0 }, { "word": "さ", "start": 21.22, "end": 21.38, "probability": 1.0 }, { "word": "れて", "start": 21.38, "end": 21.52, "probability": 1.0 }, { "word": "います", "start": 21.52, "end": 21.9, "probability": 0.99951171875 } ] }, { "id": 6, "seek": 2190, "start": 22.180000000000003, "end": 25.5, "text": "スモーキングが コーヒベントリー", "tokens": [ 50365, 9550, 29183, 3384, 15535, 43017, 5142, 220, 18066, 15266, 240, 45290, 29971, 12376, 3384, 50565 ], "temperature": 0.0, "avg_logprob": -0.7408057942109949, "compression_ratio": 0.9019607843137255, "no_speech_prob": 0.07329156249761581, "words": [ { "word": "ス", "start": 22.180000000000003, "end": 22.42, "probability": 0.1824951171875 }, { "word": "モ", "start": 22.42, "end": 22.52, "probability": 0.9423828125 }, { "word": "ー", "start": 22.52, "end": 22.66, "probability": 0.890625 }, { "word": "キ", "start": 22.66, "end": 22.68, "probability": 0.990234375 }, { "word": "ング", "start": 22.68, "end": 22.76, "probability": 0.9892578125 }, { "word": "が", "start": 22.76, "end": 22.88, "probability": 0.515625 }, { "word": " コ", "start": 23.72, "end": 23.84, "probability": 0.1697998046875 }, { "word": "ーヒ", "start": 23.84, "end": 24.0, "probability": 0.5673828125 }, { "word": "ベ", "start": 24.0, "end": 24.0, "probability": 0.1234130859375 }, { "word": "ント", "start": 24.0, "end": 24.1, "probability": 0.60986328125 }, { "word": "リ", "start": 24.1, "end": 24.1, "probability": 0.9990234375 }, { "word": "ー", "start": 24.1, "end": 25.5, "probability": 0.99951171875 } ] } ], "language": "ja", "duration": 26.6 }, "status": "COMPLETE", "rejected_alternative": { "text": "ベルト着用サイン点灯中。並びにお座りの間は常にシートベルトを締めください。 Please keep your seatbelt fastened whenever you are seated.外す場合は金具の片方を持ち上げてください。 To unfasten, raise the flap.この便は全席禁煙です。化粧室でのタバコは航空法で禁止されています。 Smoking is prohibited in the cabin, including the lavatory.携帯", "chunks": [ { "text": "ベルト着用サイン点灯中。並びにお座りの間は常にシートベルトを締めください。 Please keep your seatbelt fastened whenever you are seated.外す場合は金具の片方を持ち上げてください。 To unfasten, raise the flap.この便は全席禁煙です。化粧室でのタバコは航空法で禁止されています。 Smoking is prohibited in the cabin, including the lavatory.携帯", "start": 0.0, "end": 26.6, "chunk_index": 0, "language": "Japanese", "finish_reason": "eos", "truncated": false, "generated_tokens": 103, "max_new_tokens": 352 } ], "detected_language": "Japanese", "segments": [ { "start": 223.0, "end": 249.6, "text": "ベルト着用サイン点灯中。並びにお座りの間は常にシートベルトを締めください。 Please keep your seatbelt fastened whenever you are seated.外す場合は金具の片方を持ち上げてください。 To unfasten, raise the flap.この便は全席禁煙です。化粧室でのタバコは航空法で禁止されています。 Smoking is prohibited in the cabin, including the lavatory.携帯" } ], "start": 223.0, "end": 249.60000000000002, "engine": "mlx-qwen3-asr", "language": "ja" } }, "raw_word_timestamps": [ { "word": "携", "start": 245.22, "end": 245.54, "probability": 0.91943359375 }, { "word": "帯", "start": 245.54, "end": 245.64, "probability": 0.9990234375 }, { "word": "の", "start": 245.64, "end": 245.86, "probability": 0.196044921875 }, { "word": "タ", "start": 245.86, "end": 246.0, "probability": 0.2049560546875 }, { "word": "バ", "start": 246.0, "end": 247.0, "probability": 0.984375 }, { "word": "コ", "start": 247.0, "end": 247.22, "probability": 0.998046875 }, { "word": "は", "start": 247.22, "end": 247.42000000000002, "probability": 0.466552734375 }, { "word": "ス", "start": 247.42000000000002, "end": 247.42000000000002, "probability": 0.0271759033203125 }, { "word": "テ", "start": 247.42000000000002, "end": 247.52, "probability": 0.045166015625 }, { "word": "ー", "start": 247.52, "end": 247.57999999999998, "probability": 0.209716796875 }, { "word": "キ", "start": 247.57999999999998, "end": 247.7, "probability": 0.2021484375 }, { "word": "の", "start": 247.7, "end": 247.72, "probability": 0.11944580078125 }, { "word": "タ", "start": 247.72, "end": 248.0, "probability": 0.07244873046875 }, { "word": "バ", "start": 248.0, "end": 248.06, "probability": 0.93359375 }, { "word": "コ", "start": 248.06, "end": 248.48, "probability": 0.994140625 }, { "word": "です", "start": 248.48, "end": 248.5, "probability": 0.10516357421875 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "携帯のタバコはスモーキングです", "review_status": "BEST_SUPPORTED_WITH_REVIEW", "classification": "BEST_SUPPORTED_WITH_REVIEW", "confidence": 0.85, "correction_reason": "Independent ASR 'スモーキング' and context '禁煙' support 'smoking' over 'steak'." }, { "start": 249.11999999999998, "end": 253.42, "text": "携帯電話などの電波を発する状態にある機器はご使用になれません", "raw_asr_text": "携帯電話などの電波を発する状態にある機器はご使用になれません", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 14, "words": [ { "word": "携", "start": 249.11999999999998, "end": 249.43999999999997, "probability": 0.91796875 }, { "word": "帯", "start": 249.43999999999997, "end": 249.61999999999998, "probability": 1.0 }, { "word": "電", "start": 249.61999999999998, "end": 249.77999999999997, "probability": 0.998046875 }, { "word": "話", "start": 249.77999999999997, "end": 249.93999999999997, "probability": 1.0 }, { "word": "な", "start": 249.93999999999997, "end": 250.04, "probability": 0.98486328125 }, { "word": "ど", "start": 250.04, "end": 250.17999999999998, "probability": 0.9990234375 }, { "word": "の", "start": 250.17999999999998, "end": 250.32, "probability": 0.99609375 }, { "word": "電", "start": 250.32, "end": 250.54, "probability": 0.99072265625 }, { "word": "波", "start": 250.54, "end": 250.71999999999997, "probability": 0.99853515625 }, { "word": "を", "start": 250.71999999999997, "end": 250.82, "probability": 0.9970703125 }, { "word": "発", "start": 250.82, "end": 250.95999999999998, "probability": 0.9970703125 }, { "word": "する", "start": 250.95999999999998, "end": 251.17999999999998, "probability": 0.998046875 }, { "word": "状", "start": 251.17999999999998, "end": 251.42, "probability": 0.9677734375 }, { "word": "態", "start": 251.42, "end": 251.55999999999997, "probability": 0.99951171875 }, { "word": "に", "start": 251.55999999999997, "end": 251.7, "probability": 0.9873046875 }, { "word": "ある", "start": 251.7, "end": 251.83999999999997, "probability": 0.99365234375 }, { "word": "機", "start": 251.83999999999997, "end": 251.99999999999997, "probability": 0.986328125 }, { "word": "器", "start": 251.99999999999997, "end": 252.14, "probability": 0.9990234375 }, { "word": "は", "start": 252.14, "end": 252.26, "probability": 0.99658203125 }, { "word": "ご", "start": 252.26, "end": 252.48, "probability": 0.84375 }, { "word": "使", "start": 252.48, "end": 252.61999999999998, "probability": 0.99853515625 }, { "word": "用", "start": 252.61999999999998, "end": 252.76, "probability": 1.0 }, { "word": "にな", "start": 252.76, "end": 252.98, "probability": 0.986328125 }, { "word": "れ", "start": 252.98, "end": 253.07999999999998, "probability": 0.9990234375 }, { "word": "ません", "start": 253.07999999999998, "end": 253.42, "probability": 0.9990234375 } ], "asr_metrics": { "avg_logprob": -0.09470450250725997, "compression_ratio": 1.2857142857142858, "no_speech_prob": 1.885961872782893e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "携帯電話などの電波を発する状態にある機器はご使用になれません電波を発しない設定に切り替えていただくか電源をお切りください" }, { "language": "en", "text": "You can use the device to get a radio radio device. You can use the device to get a radio radio device. You can switch to the device to get a radio radio device. You can switch to the airplane mode or turn them off." } ], "id": 44, "review_annotation": "", "row_id": 44, "bounded_row_id": 44, "parent_row_ids": [ 44 ], "parent_row_id": 44, "packet_id": 14, "LID_evidence": [ { "region_id": 35, "probabilities": { "ja": 0.1856426125013968, "zh": 0.004833114871871658, "en": 0.38395933059456927 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 35 ], "language_confidence": 0.1856426125013968, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 248.29999999999998, "packet_end": 266.40000000000003, "raw_text": "携帯電話などの電波を発する状態にある機器はご使用になれません", "selected_text": "携帯電話などの電波を発する状態にある機器はご使用になれません", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 44 ], "packet_id": 14, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "携帯電話などの電波を発する状態にある機器はご使用になれません", "after": "携帯電話などの電波を発する状態にある機器はご使用になれません", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Clear standard instruction, matches ASR and context perfectly." } ], "language": "ja", "packet_ids": [ 14 ], "index": 44, "source_text": "携帯電話などの電波を発する状態にある機器はご使用になれません", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Clear standard instruction, matches ASR and context perfectly." }, { "start": 253.42, "end": 262.5, "text": "電波を発しない設定に切り替えていただくか電源をお切りください", "raw_asr_text": "電波を発しない設定に切り替えていただくか電源をお切りください", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 14, "words": [ { "word": "電", "start": 253.42, "end": 253.83999999999997, "probability": 0.95458984375 }, { "word": "波", "start": 253.83999999999997, "end": 258.85999999999996, "probability": 0.9951171875 }, { "word": "を", "start": 258.85999999999996, "end": 258.94, "probability": 0.99365234375 }, { "word": "発", "start": 258.94, "end": 259.12, "probability": 0.99951171875 }, { "word": "し", "start": 259.12, "end": 259.24, "probability": 0.994140625 }, { "word": "ない", "start": 259.24, "end": 259.35999999999996, "probability": 0.9990234375 }, { "word": "設", "start": 259.35999999999996, "end": 259.59999999999997, "probability": 0.998046875 }, { "word": "定", "start": 259.59999999999997, "end": 259.76, "probability": 0.9990234375 }, { "word": "に", "start": 259.76, "end": 259.88, "probability": 0.99951171875 }, { "word": "切", "start": 259.88, "end": 260.03999999999996, "probability": 0.9970703125 }, { "word": "り", "start": 260.03999999999996, "end": 260.12, "probability": 0.99853515625 }, { "word": "替", "start": 260.12, "end": 260.24, "probability": 1.0 }, { "word": "えて", "start": 260.24, "end": 260.4, "probability": 0.99951171875 }, { "word": "いただ", "start": 260.4, "end": 260.64, "probability": 0.9970703125 }, { "word": "く", "start": 260.64, "end": 260.78, "probability": 0.99951171875 }, { "word": "か", "start": 260.78, "end": 260.94, "probability": 0.99951171875 }, { "word": "電", "start": 260.94, "end": 261.41999999999996, "probability": 0.7646484375 }, { "word": "源", "start": 261.41999999999996, "end": 261.65999999999997, "probability": 0.99951171875 }, { "word": "を", "start": 261.65999999999997, "end": 261.82, "probability": 0.99365234375 }, { "word": "お", "start": 261.82, "end": 261.94, "probability": 0.8115234375 }, { "word": "切", "start": 261.94, "end": 262.02, "probability": 0.99853515625 }, { "word": "り", "start": 262.02, "end": 262.12, "probability": 1.0 }, { "word": "ください", "start": 262.12, "end": 262.5, "probability": 0.99658203125 } ], "asr_metrics": { "avg_logprob": -0.09470450250725997, "compression_ratio": 1.2857142857142858, "no_speech_prob": 1.885961872782893e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "携帯電話などの電波を発する状態にある機器はご使用になれません電波を発しない設定に切り替えていただくか電源をお切りください" }, { "language": "en", "text": "You can use the device to get a radio radio device. You can use the device to get a radio radio device. You can switch to the device to get a radio radio device. You can switch to the airplane mode or turn them off." } ], "id": 45, "review_annotation": "", "row_id": 45, "bounded_row_id": 45, "parent_row_ids": [ 45 ], "parent_row_id": 45, "packet_id": 14, "LID_evidence": [ { "region_id": 35, "probabilities": { "ja": 0.1856426125013968, "zh": 0.004833114871871658, "en": 0.38395933059456927 }, "status": "UNCERTAIN" }, { "region_id": 36, "probabilities": { "ja": 0.22463868624569538, "zh": 0.004110482034351056, "en": 0.4057679068918029 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 35, 36 ], "language_confidence": 0.22463868624569538, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 248.29999999999998, "packet_end": 266.40000000000003, "raw_text": "電波を発しない設定に切り替えていただくか電源をお切りください", "selected_text": "電波を発しない設定に切り替えていただくか電源をお切りください", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 45 ], "packet_id": 14, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "電波を発しない設定に切り替えていただくか電源をお切りください", "after": "電波を発しない設定に切り替えていただくか電源をお切りください", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard instruction regarding electronic devices, matches ASR and context." } ], "language": "ja", "packet_ids": [ 14 ], "index": 45, "source_text": "電波を発しない設定に切り替えていただくか電源をお切りください", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Standard instruction regarding electronic devices, matches ASR and context." }, { "start": 267.96, "end": 270.34, "text": "次に、非常用設備をご案内します", "raw_asr_text": "次に非常用設備をご案内します", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 15, "words": [ { "word": "次", "start": 267.96, "end": 268.23999999999995, "probability": 0.98974609375 }, { "word": "に", "start": 268.23999999999995, "end": 268.52, "probability": 0.99951171875 }, { "word": "非常", "start": 268.52, "end": 268.82, "probability": 0.8798828125 }, { "word": "用", "start": 268.82, "end": 269.05999999999995, "probability": 0.98974609375 }, { "word": "設", "start": 269.05999999999995, "end": 269.26, "probability": 0.9912109375 }, { "word": "備", "start": 269.26, "end": 269.43999999999994, "probability": 1.0 }, { "word": "を", "start": 269.43999999999994, "end": 269.55999999999995, "probability": 1.0 }, { "word": "ご", "start": 269.55999999999995, "end": 269.67999999999995, "probability": 0.99951171875 }, { "word": "案", "start": 269.67999999999995, "end": 269.84, "probability": 0.99951171875 }, { "word": "内", "start": 269.84, "end": 269.97999999999996, "probability": 1.0 }, { "word": "します", "start": 269.97999999999996, "end": 270.34, "probability": 1.0 } ], "asr_metrics": { "avg_logprob": -0.13412359677828276, "compression_ratio": 1.2516556291390728, "no_speech_prob": 7.909142091255816e-12 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "次に非常用設備をご案内します酸素が必要な状態になると酸素マスクが降りてきますマスクを強く引き寄せ鼻と口に当てて呼吸してください" }, { "language": "en", "text": "Next, the equipment will be used to be a mask. If you need a mask, you can get a mask. Pull the mask firmly towards you. Place it over your nose and mouth and breathe normally." } ], "id": 46, "review_annotation": "", "row_id": 46, "bounded_row_id": 46, "parent_row_ids": [ 46 ], "parent_row_id": 46, "packet_id": 15, "LID_evidence": [ { "region_id": 37, "probabilities": { "ja": 0.44339166678400943, "zh": 0.0040625348313672776, "en": 0.47492351576996344 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 37 ], "language_confidence": 0.44339166678400943, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 267.09999999999997, "packet_end": 290.3, "raw_text": "次に非常用設備をご案内します", "selected_text": "次に、非常用設備をご案内します", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 46 ], "packet_id": 15, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "次に、非常用設備をご案内します", "after": "次に、非常用設備をご案内します", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard transition phrase, matches ASR and context." } ], "language": "ja", "deterministic_correction_applied": true, "packet_ids": [ 15 ], "index": 46, "source_text": "次に、非常用設備をご案内します", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Standard transition phrase, matches ASR and context." }, { "start": 274.37999999999994, "end": 277.73999999999995, "text": "酸素が必要な状態になると酸素マスクが降りてきます", "raw_asr_text": "酸素が必要な状態になると酸素マスクが降りてきます", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 15, "words": [ { "word": "酸", "start": 274.37999999999994, "end": 274.65999999999997, "probability": 0.537109375 }, { "word": "素", "start": 274.65999999999997, "end": 274.79999999999995, "probability": 0.99951171875 }, { "word": "が", "start": 274.79999999999995, "end": 274.9, "probability": 0.99658203125 }, { "word": "必", "start": 274.9, "end": 275.05999999999995, "probability": 0.99951171875 }, { "word": "要", "start": 275.05999999999995, "end": 275.23999999999995, "probability": 1.0 }, { "word": "な", "start": 275.23999999999995, "end": 275.35999999999996, "probability": 0.99951171875 }, { "word": "状", "start": 275.35999999999996, "end": 275.55999999999995, "probability": 0.9990234375 }, { "word": "態", "start": 275.55999999999995, "end": 275.7, "probability": 0.99951171875 }, { "word": "にな", "start": 275.7, "end": 275.91999999999996, "probability": 1.0 }, { "word": "る", "start": 275.91999999999996, "end": 276.03999999999996, "probability": 0.99951171875 }, { "word": "と", "start": 276.03999999999996, "end": 276.2, "probability": 1.0 }, { "word": "酸", "start": 276.2, "end": 276.64, "probability": 0.955078125 }, { "word": "素", "start": 276.64, "end": 276.78, "probability": 1.0 }, { "word": "マ", "start": 276.78, "end": 276.9, "probability": 0.99951171875 }, { "word": "ス", "start": 276.9, "end": 276.97999999999996, "probability": 1.0 }, { "word": "ク", "start": 276.97999999999996, "end": 277.03999999999996, "probability": 1.0 }, { "word": "が", "start": 277.03999999999996, "end": 277.17999999999995, "probability": 0.99951171875 }, { "word": "降", "start": 277.17999999999995, "end": 277.29999999999995, "probability": 0.89013671875 }, { "word": "り", "start": 277.29999999999995, "end": 277.34, "probability": 0.9990234375 }, { "word": "て", "start": 277.34, "end": 277.43999999999994, "probability": 1.0 }, { "word": "きます", "start": 277.43999999999994, "end": 277.73999999999995, "probability": 0.99951171875 } ], "asr_metrics": { "avg_logprob": -0.13412359677828276, "compression_ratio": 1.2516556291390728, "no_speech_prob": 7.909142091255816e-12 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "次に非常用設備をご案内します酸素が必要な状態になると酸素マスクが降りてきますマスクを強く引き寄せ鼻と口に当てて呼吸してください" }, { "language": "en", "text": "Next, the equipment will be used to be a mask. If you need a mask, you can get a mask. Pull the mask firmly towards you. Place it over your nose and mouth and breathe normally." } ], "id": 47, "review_annotation": "", "row_id": 47, "bounded_row_id": 47, "parent_row_ids": [ 47 ], "parent_row_id": 47, "packet_id": 15, "LID_evidence": [ { "region_id": 38, "probabilities": { "ja": 0.547731758832621, "zh": 0.004495095889271599, "en": 0.3122392449359823 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 38 ], "language_confidence": 0.547731758832621, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 267.09999999999997, "packet_end": 290.3, "raw_text": "酸素が必要な状態になると酸素マスクが降りてきます", "selected_text": "酸素が必要な状態になると酸素マスクが降りてきます", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 47 ], "packet_id": 15, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "酸素が必要な状態になると酸素マスクが降りてきます", "after": "酸素が必要な状態になると酸素マスクが降りてきます", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard safety instruction, matches ASR and context." } ], "language": "ja", "packet_ids": [ 15 ], "index": 47, "source_text": "酸素が必要な状態になると酸素マスクが降りてきます", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Standard safety instruction, matches ASR and context." }, { "start": 281.11999999999995, "end": 284.65999999999997, "text": "マスクを強く引き寄せ鼻と口に当てて呼吸してください", "raw_asr_text": "マスクを強く引き寄せ鼻と口に当てて呼吸してください", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 15, "words": [ { "word": "マ", "start": 281.11999999999995, "end": 281.4, "probability": 0.931640625 }, { "word": "ス", "start": 281.4, "end": 281.49999999999994, "probability": 1.0 }, { "word": "ク", "start": 281.49999999999994, "end": 281.59999999999997, "probability": 1.0 }, { "word": "を", "start": 281.59999999999997, "end": 281.7, "probability": 0.9990234375 }, { "word": "強", "start": 281.7, "end": 281.87999999999994, "probability": 0.9990234375 }, { "word": "く", "start": 281.87999999999994, "end": 282.03999999999996, "probability": 1.0 }, { "word": "引", "start": 282.03999999999996, "end": 282.17999999999995, "probability": 0.99951171875 }, { "word": "き", "start": 282.17999999999995, "end": 282.28, "probability": 1.0 }, { "word": "寄", "start": 282.28, "end": 282.35999999999996, "probability": 0.99951171875 }, { "word": "せ", "start": 282.35999999999996, "end": 282.53999999999996, "probability": 1.0 }, { "word": "鼻", "start": 282.53999999999996, "end": 282.99999999999994, "probability": 0.9306640625 }, { "word": "と", "start": 282.99999999999994, "end": 283.11999999999995, "probability": 0.9990234375 }, { "word": "口", "start": 283.11999999999995, "end": 283.28, "probability": 0.99951171875 }, { "word": "に", "start": 283.28, "end": 283.41999999999996, "probability": 1.0 }, { "word": "当", "start": 283.41999999999996, "end": 283.52, "probability": 0.98681640625 }, { "word": "て", "start": 283.52, "end": 283.59999999999997, "probability": 1.0 }, { "word": "て", "start": 283.59999999999997, "end": 283.73999999999995, "probability": 1.0 }, { "word": "呼", "start": 283.73999999999995, "end": 283.93999999999994, "probability": 0.9990234375 }, { "word": "吸", "start": 283.93999999999994, "end": 284.08, "probability": 0.99951171875 }, { "word": "して", "start": 284.08, "end": 284.26, "probability": 0.9990234375 }, { "word": "ください", "start": 284.26, "end": 284.65999999999997, "probability": 0.99951171875 } ], "asr_metrics": { "avg_logprob": -0.13412359677828276, "compression_ratio": 1.2516556291390728, "no_speech_prob": 7.909142091255816e-12 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "次に非常用設備をご案内します酸素が必要な状態になると酸素マスクが降りてきますマスクを強く引き寄せ鼻と口に当てて呼吸してください" }, { "language": "en", "text": "Next, the equipment will be used to be a mask. If you need a mask, you can get a mask. Pull the mask firmly towards you. Place it over your nose and mouth and breathe normally." } ], "id": 48, "review_annotation": "", "row_id": 48, "bounded_row_id": 48, "parent_row_ids": [ 48 ], "parent_row_id": 48, "packet_id": 15, "LID_evidence": [ { "region_id": 39, "probabilities": { "ja": 0.3725379082773413, "zh": 0.0032359669608662705, "en": 0.4736822858186705 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 39 ], "language_confidence": 0.3725379082773413, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 267.09999999999997, "packet_end": 290.3, "raw_text": "マスクを強く引き寄せ鼻と口に当てて呼吸してください", "selected_text": "マスクを強く引き寄せ鼻と口に当てて呼吸してください", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 48 ], "packet_id": 15, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "マスクを強く引き寄せ鼻と口に当てて呼吸してください", "after": "マスクを強く引き寄せ鼻と口に当てて呼吸してください", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard safety instruction, matches ASR and context." } ], "language": "ja", "packet_ids": [ 15 ], "index": 48, "source_text": "マスクを強く引き寄せ鼻と口に当てて呼吸してください", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Standard safety instruction, matches ASR and context." }, { "start": 290.1, "end": 292.98, "text": "ゴム紐は頭にかけ長さを調整してください", "raw_asr_text": "ゴム紐は頭にかけ長さを調整してください", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 16, "words": [], "asr_metrics": { "avg_logprob": -0.1461583375930786, "compression_ratio": 1.25, "no_speech_prob": 6.841906815407972e-12 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "ゴム紐は頭にかけ長さを調整してください緊急の場合は乗務員の指示に従い衝撃に備える姿勢をとってください非常口は機体の前方、中央、後方にあります" }, { "language": "en", "text": "Goom himo is head and adjust the length of the elastic band. In an emergency, brace for impact. The front door is located in the front, center and rear of the cabin." } ], "id": 49, "review_annotation": "", "row_id": 49, "bounded_row_id": 49, "parent_row_ids": [ 49 ], "parent_row_id": 49, "packet_id": 16, "LID_evidence": [ { "region_id": 40, "probabilities": { "ja": 0.3755295261761339, "zh": 0.006122436530272206, "en": 0.45391911670805946 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 40 ], "language_confidence": 0.3755295261761339, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 289.3, "packet_end": 316.1, "raw_text": "ゴム紐は頭にかけ長さを調整してください", "selected_text": "ゴム紐は頭にかけ長さを調整してください", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 49 ], "packet_id": 16, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "ゴム紐は頭にかけ長さを調整してください", "after": "ゴム紐は頭にかけ長さを調整してください", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard safety instruction, matches ASR and context." } ], "language": "ja", "packet_ids": [ 16 ], "index": 49, "source_text": "ゴム紐は頭にかけ長さを調整してください", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Standard safety instruction, matches ASR and context.", "raw_word_timestamps": [ { "word": "ゴ", "start": 290.02000000000004, "end": 290.38, "probability": 0.96484375 }, { "word": "ム", "start": 290.38, "end": 290.52000000000004, "probability": 0.99951171875 }, { "word": "紐", "start": 290.52000000000004, "end": 290.66, "probability": 0.8759765625 }, { "word": "は", "start": 290.66, "end": 290.84000000000003, "probability": 0.93603515625 }, { "word": "頭", "start": 290.84000000000003, "end": 291.06, "probability": 0.9970703125 }, { "word": "に", "start": 291.06, "end": 291.24, "probability": 1.0 }, { "word": "か", "start": 291.24, "end": 291.36, "probability": 0.78125 }, { "word": "け", "start": 291.36, "end": 291.5, "probability": 0.998046875 }, { "word": "長", "start": 291.5, "end": 291.78000000000003, "probability": 0.62255859375 }, { "word": "さ", "start": 291.78000000000003, "end": 291.94, "probability": 1.0 }, { "word": "を", "start": 291.94, "end": 292.04, "probability": 1.0 }, { "word": "調", "start": 292.04, "end": 292.18, "probability": 0.99951171875 }, { "word": "整", "start": 292.18, "end": 292.40000000000003, "probability": 0.99951171875 }, { "word": "して", "start": 292.40000000000003, "end": 292.56, "probability": 0.9990234375 }, { "word": "ください", "start": 292.56, "end": 292.98, "probability": 0.99755859375 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 292.98, "end": 302.2, "text": "緊急の場合は乗務員の指示に従い衝撃に備える姿勢をとってください", "raw_asr_text": "緊急の場合は乗務員の指示に従い衝撃に備える姿勢をとってください", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 16, "words": [ { "word": "緊", "start": 292.98, "end": 293.52000000000004, "probability": 0.37060546875 }, { "word": "急", "start": 293.52000000000004, "end": 297.94, "probability": 1.0 }, { "word": "の", "start": 297.94, "end": 298.1, "probability": 0.99462890625 }, { "word": "場", "start": 298.1, "end": 298.28000000000003, "probability": 0.99951171875 }, { "word": "合", "start": 298.28000000000003, "end": 298.38, "probability": 1.0 }, { "word": "は", "start": 298.38, "end": 298.56, "probability": 1.0 }, { "word": "乗", "start": 298.56, "end": 299.0, "probability": 0.98193359375 }, { "word": "務", "start": 299.0, "end": 299.14, "probability": 0.9990234375 }, { "word": "員", "start": 299.14, "end": 299.24, "probability": 0.99951171875 }, { "word": "の", "start": 299.24, "end": 299.36, "probability": 1.0 }, { "word": "指", "start": 299.36, "end": 299.48, "probability": 1.0 }, { "word": "示", "start": 299.48, "end": 299.6, "probability": 0.99951171875 }, { "word": "に", "start": 299.6, "end": 299.72, "probability": 1.0 }, { "word": "従", "start": 299.72, "end": 299.92, "probability": 0.9990234375 }, { "word": "い", "start": 299.92, "end": 300.12, "probability": 0.99951171875 }, { "word": "衝", "start": 300.12, "end": 300.42, "probability": 0.935546875 }, { "word": "撃", "start": 300.42, "end": 300.62, "probability": 1.0 }, { "word": "に", "start": 300.62, "end": 300.74, "probability": 0.9990234375 }, { "word": "備", "start": 300.74, "end": 300.92, "probability": 0.99951171875 }, { "word": "える", "start": 300.92, "end": 301.14, "probability": 0.998046875 }, { "word": "姿", "start": 301.14, "end": 301.3, "probability": 1.0 }, { "word": "勢", "start": 301.3, "end": 301.40000000000003, "probability": 0.99951171875 }, { "word": "を", "start": 301.40000000000003, "end": 301.52000000000004, "probability": 1.0 }, { "word": "と", "start": 301.52000000000004, "end": 301.64, "probability": 0.982421875 }, { "word": "って", "start": 301.64, "end": 301.76, "probability": 1.0 }, { "word": "ください", "start": 301.76, "end": 302.2, "probability": 1.0 } ], "asr_metrics": { "avg_logprob": -0.1461583375930786, "compression_ratio": 1.25, "no_speech_prob": 6.841906815407972e-12 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "ゴム紐は頭にかけ長さを調整してください緊急の場合は乗務員の指示に従い衝撃に備える姿勢をとってください非常口は機体の前方、中央、後方にあります" }, { "language": "en", "text": "Goom himo is head and adjust the length of the elastic band. In an emergency, brace for impact. The front door is located in the front, center and rear of the cabin." } ], "id": 50, "review_annotation": "", "row_id": 50, "bounded_row_id": 50, "parent_row_ids": [ 50 ], "parent_row_id": 50, "packet_id": 16, "LID_evidence": [ { "region_id": 40, "probabilities": { "ja": 0.3755295261761339, "zh": 0.006122436530272206, "en": 0.45391911670805946 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 40 ], "language_confidence": 0.3755295261761339, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 289.3, "packet_end": 316.1, "raw_text": "緊急の場合は乗務員の指示に従い衝撃に備える姿勢をとってください", "selected_text": "緊急の場合は乗務員の指示に従い衝撃に備える姿勢をとってください", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 50 ], "packet_id": 16, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "緊急の場合は乗務員の指示に従い衝撃に備える姿勢をとってください", "after": "緊急の場合は乗務員の指示に従い衝撃に備える姿勢をとってください", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard safety announcement phrase matches raw ASR and context perfectly." } ], "language": "ja", "packet_ids": [ 16 ], "index": 50, "source_text": "緊急の場合は乗務員の指示に従い衝撃に備える姿勢をとってください", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard safety announcement phrase matches raw ASR and context perfectly." }, { "start": 306.84000000000003, "end": 310.52, "text": "非常口は機体の前方、中央、後方にあります", "raw_asr_text": "非常口は機体の前方、中央、後方にあります", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 16, "words": [ { "word": "非常", "start": 306.84000000000003, "end": 307.2, "probability": 0.986328125 }, { "word": "口", "start": 307.2, "end": 307.42, "probability": 0.9970703125 }, { "word": "は", "start": 307.42, "end": 307.7, "probability": 0.99951171875 }, { "word": "機", "start": 307.7, "end": 308.04, "probability": 0.990234375 }, { "word": "体", "start": 308.04, "end": 308.18, "probability": 0.99755859375 }, { "word": "の", "start": 308.18, "end": 308.34000000000003, "probability": 0.99951171875 }, { "word": "前", "start": 308.34000000000003, "end": 308.52, "probability": 0.99951171875 }, { "word": "方、", "start": 308.52, "end": 308.76, "probability": 0.99853515625 }, { "word": "中", "start": 308.98, "end": 309.16, "probability": 0.9990234375 }, { "word": "央、", "start": 309.16, "end": 309.46000000000004, "probability": 0.9912109375 }, { "word": "後", "start": 309.58000000000004, "end": 309.76, "probability": 0.99951171875 }, { "word": "方", "start": 309.76, "end": 309.96000000000004, "probability": 1.0 }, { "word": "に", "start": 309.96000000000004, "end": 310.1, "probability": 1.0 }, { "word": "あります", "start": 310.1, "end": 310.52, "probability": 1.0 } ], "asr_metrics": { "avg_logprob": -0.1461583375930786, "compression_ratio": 1.25, "no_speech_prob": 6.841906815407972e-12 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "ゴム紐は頭にかけ長さを調整してください緊急の場合は乗務員の指示に従い衝撃に備える姿勢をとってください非常口は機体の前方、中央、後方にあります" }, { "language": "en", "text": "Goom himo is head and adjust the length of the elastic band. In an emergency, brace for impact. The front door is located in the front, center and rear of the cabin." } ], "id": 51, "review_annotation": "", "row_id": 51, "bounded_row_id": 51, "parent_row_ids": [ 51 ], "parent_row_id": 51, "packet_id": 16, "LID_evidence": [ { "region_id": 41, "probabilities": { "ja": 0.3547765832062235, "zh": 0.002155138418696451, "en": 0.5249958439072704 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 41 ], "language_confidence": 0.3547765832062235, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 289.3, "packet_end": 316.1, "raw_text": "非常口は機体の前方、中央、後方にあります", "selected_text": "非常口は機体の前方、中央、後方にあります", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 51 ], "packet_id": 16, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "非常口は機体の前方、中央、後方にあります", "after": "非常口は機体の前方、中央、後方にあります", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard safety announcement phrase matches raw ASR and context perfectly." } ], "language": "ja", "packet_ids": [ 16 ], "index": 51, "source_text": "非常口は機体の前方、中央、後方にあります", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard safety announcement phrase matches raw ASR and context perfectly." }, { "start": 316.2, "end": 318.41999999999996, "text": "お近くの非常口をお確かめください", "raw_asr_text": "お近くの非常口をお確かめください", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 17, "words": [], "asr_metrics": { "avg_logprob": -0.14915558695793152, "compression_ratio": 1.2887323943661972, "no_speech_prob": 1.4455328427309677e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "お近くの非常口をお確かめください脱出の際には経路と非常口を示す明かりが点灯します乗務員の指示に従い行動してください手荷物は" } ], "id": 52, "review_annotation": "", "row_id": 52, "bounded_row_id": 52, "parent_row_ids": [ 52 ], "parent_row_id": 52, "packet_id": 17, "LID_evidence": [ { "region_id": 42, "probabilities": { "ja": 0.4901382021304959, "zh": 0.0033548629878539403, "en": 0.30756183776840634 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 42 ], "language_confidence": 0.4901382021304959, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 315.4, "packet_end": 334.40000000000003, "raw_text": "お近くの非常口をお確かめください", "selected_text": "お近くの非常口をお確かめください", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 52 ], "packet_id": 17, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "お近くの非常口をお確かめください", "after": "お近くの非常口をお確かめください", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard safety announcement phrase matches raw ASR and context perfectly." } ], "language": "ja", "packet_ids": [ 17 ], "index": 52, "source_text": "お近くの非常口をお確かめください", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard safety announcement phrase matches raw ASR and context perfectly.", "raw_word_timestamps": [ { "word": "お", "start": 316.15999999999997, "end": 316.47999999999996, "probability": 0.9453125 }, { "word": "近", "start": 316.47999999999996, "end": 316.64, "probability": 0.9970703125 }, { "word": "く", "start": 316.64, "end": 316.78, "probability": 0.99853515625 }, { "word": "の", "start": 316.78, "end": 316.9, "probability": 0.99951171875 }, { "word": "非常", "start": 316.9, "end": 317.09999999999997, "probability": 0.99853515625 }, { "word": "口", "start": 317.09999999999997, "end": 317.32, "probability": 0.9990234375 }, { "word": "を", "start": 317.32, "end": 317.56, "probability": 0.99951171875 }, { "word": "お", "start": 317.56, "end": 317.71999999999997, "probability": 0.9814453125 }, { "word": "確", "start": 317.71999999999997, "end": 317.82, "probability": 0.99951171875 }, { "word": "か", "start": 317.82, "end": 317.97999999999996, "probability": 0.99951171875 }, { "word": "め", "start": 317.97999999999996, "end": 318.08, "probability": 1.0 }, { "word": "ください", "start": 318.08, "end": 318.41999999999996, "probability": 0.998046875 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 320.56, "end": 324.56, "text": "脱出の際には経路と非常口を示す明かりが点灯します", "raw_asr_text": "脱出の際には経路と非常口を示す明かりが点灯します", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 17, "words": [ { "word": "脱", "start": 320.56, "end": 320.88, "probability": 0.64208984375 }, { "word": "出", "start": 320.88, "end": 321.06, "probability": 0.9990234375 }, { "word": "の", "start": 321.06, "end": 321.2, "probability": 0.998046875 }, { "word": "際", "start": 321.2, "end": 321.4, "probability": 0.99951171875 }, { "word": "には", "start": 321.4, "end": 321.67999999999995, "probability": 1.0 }, { "word": "経", "start": 321.67999999999995, "end": 322.12, "probability": 0.88525390625 }, { "word": "路", "start": 322.12, "end": 322.21999999999997, "probability": 1.0 }, { "word": "と", "start": 322.21999999999997, "end": 322.34, "probability": 0.99951171875 }, { "word": "非常", "start": 322.34, "end": 322.58, "probability": 0.99951171875 }, { "word": "口", "start": 322.58, "end": 322.79999999999995, "probability": 1.0 }, { "word": "を", "start": 322.79999999999995, "end": 322.96, "probability": 1.0 }, { "word": "示", "start": 322.96, "end": 323.12, "probability": 0.99951171875 }, { "word": "す", "start": 323.12, "end": 323.35999999999996, "probability": 1.0 }, { "word": "明", "start": 323.35999999999996, "end": 323.53999999999996, "probability": 0.9765625 }, { "word": "か", "start": 323.53999999999996, "end": 323.65999999999997, "probability": 0.9951171875 }, { "word": "り", "start": 323.65999999999997, "end": 323.71999999999997, "probability": 1.0 }, { "word": "が", "start": 323.71999999999997, "end": 323.85999999999996, "probability": 0.99951171875 }, { "word": "点", "start": 323.85999999999996, "end": 323.97999999999996, "probability": 0.9990234375 }, { "word": "灯", "start": 323.97999999999996, "end": 324.17999999999995, "probability": 0.99951171875 }, { "word": "します", "start": 324.17999999999995, "end": 324.56, "probability": 0.99951171875 } ], "asr_metrics": { "avg_logprob": -0.14915558695793152, "compression_ratio": 1.2887323943661972, "no_speech_prob": 1.4455328427309677e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "お近くの非常口をお確かめください脱出の際には経路と非常口を示す明かりが点灯します乗務員の指示に従い行動してください手荷物は" } ], "id": 53, "review_annotation": "", "row_id": 53, "bounded_row_id": 53, "parent_row_ids": [ 53 ], "parent_row_id": 53, "packet_id": 17, "LID_evidence": [ { "region_id": 42, "probabilities": { "ja": 0.4901382021304959, "zh": 0.0033548629878539403, "en": 0.30756183776840634 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 42 ], "language_confidence": 0.4901382021304959, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 315.4, "packet_end": 334.40000000000003, "raw_text": "脱出の際には経路と非常口を示す明かりが点灯します", "selected_text": "脱出の際には経路と非常口を示す明かりが点灯します", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 53 ], "packet_id": 17, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "脱出の際には経路と非常口を示す明かりが点灯します", "after": "脱出の際には経路と非常口を示す明かりが点灯します", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard safety announcement phrase matches raw ASR and context perfectly." } ], "language": "ja", "packet_ids": [ 17 ], "index": 53, "source_text": "脱出の際には経路と非常口を示す明かりが点灯します", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard safety announcement phrase matches raw ASR and context perfectly." }, { "start": 328.21999999999997, "end": 330.62, "text": "乗務員の指示に従い行動してください", "raw_asr_text": "乗務員の指示に従い行動してください", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 17, "words": [ { "word": "乗", "start": 328.21999999999997, "end": 328.53999999999996, "probability": 0.966796875 }, { "word": "務", "start": 328.53999999999996, "end": 328.67999999999995, "probability": 0.99755859375 }, { "word": "員", "start": 328.67999999999995, "end": 328.78, "probability": 0.99951171875 }, { "word": "の", "start": 328.78, "end": 328.91999999999996, "probability": 1.0 }, { "word": "指", "start": 328.91999999999996, "end": 329.03999999999996, "probability": 1.0 }, { "word": "示", "start": 329.03999999999996, "end": 329.15999999999997, "probability": 0.99951171875 }, { "word": "に", "start": 329.15999999999997, "end": 329.28, "probability": 1.0 }, { "word": "従", "start": 329.28, "end": 329.5, "probability": 1.0 }, { "word": "い", "start": 329.5, "end": 329.67999999999995, "probability": 0.99951171875 }, { "word": "行", "start": 329.67999999999995, "end": 329.88, "probability": 0.99755859375 }, { "word": "動", "start": 329.88, "end": 330.03999999999996, "probability": 1.0 }, { "word": "して", "start": 330.03999999999996, "end": 330.21999999999997, "probability": 0.99951171875 }, { "word": "ください", "start": 330.21999999999997, "end": 330.62, "probability": 1.0 } ], "asr_metrics": { "avg_logprob": -0.14915558695793152, "compression_ratio": 1.2887323943661972, "no_speech_prob": 1.4455328427309677e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "お近くの非常口をお確かめください脱出の際には経路と非常口を示す明かりが点灯します乗務員の指示に従い行動してください手荷物は" } ], "id": 54, "review_annotation": "", "row_id": 54, "bounded_row_id": 54, "parent_row_ids": [ 54 ], "parent_row_id": 54, "packet_id": 17, "LID_evidence": [ { "region_id": 43, "probabilities": { "ja": 0.4755777734394845, "zh": 0.00378228290695309, "en": 0.3043794033117592 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 43 ], "language_confidence": 0.4755777734394845, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 315.4, "packet_end": 334.40000000000003, "raw_text": "乗務員の指示に従い行動してください", "selected_text": "乗務員の指示に従い行動してください", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 54 ], "packet_id": 17, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "乗務員の指示に従い行動してください", "after": "乗務員の指示に従い行動してください", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard safety announcement phrase matches raw ASR and context perfectly." } ], "language": "ja", "packet_ids": [ 17 ], "index": 54, "source_text": "乗務員の指示に従い行動してください", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard safety announcement phrase matches raw ASR and context perfectly." }, { "start": 333.7, "end": 336.5, "text": "手荷物は一切持たずハイヒールは脱いでください", "raw_asr_text": "手荷物は一切持たずハイヒールは脱いでください", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 18, "words": [ { "word": "手", "start": 333.7, "end": 333.97999999999996, "probability": 0.93603515625 }, { "word": "荷", "start": 333.97999999999996, "end": 334.14, "probability": 0.998046875 }, { "word": "物", "start": 334.14, "end": 334.28, "probability": 0.9990234375 }, { "word": "は", "start": 334.28, "end": 334.44, "probability": 0.99658203125 }, { "word": "一", "start": 334.44, "end": 334.58, "probability": 0.9990234375 }, { "word": "切", "start": 334.58, "end": 334.76, "probability": 0.99951171875 }, { "word": "持", "start": 334.76, "end": 334.9, "probability": 0.99755859375 }, { "word": "た", "start": 334.9, "end": 335.02, "probability": 1.0 }, { "word": "ず", "start": 335.02, "end": 335.15999999999997, "probability": 1.0 }, { "word": "ハ", "start": 335.15999999999997, "end": 335.4, "probability": 0.68603515625 }, { "word": "イ", "start": 335.4, "end": 335.46, "probability": 0.99951171875 }, { "word": "ヒ", "start": 335.46, "end": 335.59999999999997, "probability": 0.99951171875 }, { "word": "ール", "start": 335.59999999999997, "end": 335.73999999999995, "probability": 0.99951171875 }, { "word": "は", "start": 335.73999999999995, "end": 335.88, "probability": 0.998046875 }, { "word": "脱", "start": 335.88, "end": 336.0, "probability": 0.99658203125 }, { "word": "い", "start": 336.0, "end": 336.08, "probability": 0.99853515625 }, { "word": "で", "start": 336.08, "end": 336.17999999999995, "probability": 0.9990234375 }, { "word": "ください", "start": 336.17999999999995, "end": 336.5, "probability": 0.99365234375 } ], "asr_metrics": { "avg_logprob": -0.1494094887558295, "compression_ratio": 1.4492753623188406, "no_speech_prob": 1.1984709231971369e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "手荷物は一切持たずハイヒールは脱いでくださいスライドを使用する際は両腕を前に伸ばし上体を起こして滑ります援助する場合はスライドの横に立ち体を引き起こしてください立ち上がりすぐに飛行機から離れてください行きましょう。" }, { "language": "en", "text": "Leave all baggage behind and remove high-heeled shoes. Put both arms out, sit up straight and slide down. Help the others up to their feet. Move away from the aircraft immediately. immediately." } ], "id": 55, "review_annotation": "", "row_id": 55, "bounded_row_id": 55, "parent_row_ids": [ 55 ], "parent_row_id": 55, "packet_id": 18, "LID_evidence": [ { "region_id": 44, "probabilities": { "ja": 0.3560518695782601, "zh": 0.005039430592079681, "en": 0.37702566033546275 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 44 ], "language_confidence": 0.3560518695782601, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 332.9, "packet_end": 361.8, "raw_text": "手荷物は一切持たずハイヒールは脱いでください", "selected_text": "手荷物は一切持たずハイヒールは脱いでください", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 55 ], "packet_id": 18, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "手荷物は一切持たずハイヒールは脱いでください", "after": "手荷物は一切持たずハイヒールは脱いでください", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard safety announcement phrase matches raw ASR and context perfectly." } ], "language": "ja", "packet_ids": [ 18 ], "index": 55, "source_text": "手荷物は一切持たずハイヒールは脱いでください", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard safety announcement phrase matches raw ASR and context perfectly." }, { "start": 340.41999999999996, "end": 345.0, "text": "スライドを使用する際は両腕を前に伸ばし上体を起こして滑ります", "raw_asr_text": "スライドを使用する際は両腕を前に伸ばし上体を起こして滑ります", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 18, "words": [ { "word": "ス", "start": 340.41999999999996, "end": 340.7, "probability": 0.495849609375 }, { "word": "ライ", "start": 340.7, "end": 340.84, "probability": 0.99951171875 }, { "word": "ド", "start": 340.84, "end": 340.97999999999996, "probability": 0.99951171875 }, { "word": "を", "start": 340.97999999999996, "end": 341.09999999999997, "probability": 0.99853515625 }, { "word": "使", "start": 341.09999999999997, "end": 341.21999999999997, "probability": 0.99072265625 }, { "word": "用", "start": 341.21999999999997, "end": 341.34, "probability": 0.9990234375 }, { "word": "する", "start": 341.34, "end": 341.53999999999996, "probability": 1.0 }, { "word": "際", "start": 341.53999999999996, "end": 341.73999999999995, "probability": 0.99951171875 }, { "word": "は", "start": 341.73999999999995, "end": 341.97999999999996, "probability": 0.99951171875 }, { "word": "両", "start": 341.97999999999996, "end": 342.38, "probability": 0.99462890625 }, { "word": "腕", "start": 342.38, "end": 342.53999999999996, "probability": 1.0 }, { "word": "を", "start": 342.53999999999996, "end": 342.65999999999997, "probability": 1.0 }, { "word": "前", "start": 342.65999999999997, "end": 342.84, "probability": 0.99951171875 }, { "word": "に", "start": 342.84, "end": 342.96, "probability": 1.0 }, { "word": "伸", "start": 342.96, "end": 343.08, "probability": 1.0 }, { "word": "ば", "start": 343.08, "end": 343.2, "probability": 1.0 }, { "word": "し", "start": 343.2, "end": 343.38, "probability": 0.99951171875 }, { "word": "上", "start": 343.38, "end": 343.62, "probability": 0.9755859375 }, { "word": "体", "start": 343.62, "end": 343.78, "probability": 0.990234375 }, { "word": "を", "start": 343.78, "end": 343.96, "probability": 1.0 }, { "word": "起", "start": 343.96, "end": 344.09999999999997, "probability": 0.9990234375 }, { "word": "こ", "start": 344.09999999999997, "end": 344.17999999999995, "probability": 1.0 }, { "word": "して", "start": 344.17999999999995, "end": 344.34, "probability": 1.0 }, { "word": "滑", "start": 344.34, "end": 344.53999999999996, "probability": 0.9970703125 }, { "word": "ります", "start": 344.53999999999996, "end": 345.0, "probability": 0.9931640625 } ], "asr_metrics": { "avg_logprob": -0.1494094887558295, "compression_ratio": 1.4492753623188406, "no_speech_prob": 1.1984709231971369e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "手荷物は一切持たずハイヒールは脱いでくださいスライドを使用する際は両腕を前に伸ばし上体を起こして滑ります援助する場合はスライドの横に立ち体を引き起こしてください立ち上がりすぐに飛行機から離れてください行きましょう。" }, { "language": "en", "text": "Leave all baggage behind and remove high-heeled shoes. Put both arms out, sit up straight and slide down. Help the others up to their feet. Move away from the aircraft immediately. immediately." } ], "id": 56, "review_annotation": "", "row_id": 56, "bounded_row_id": 56, "parent_row_ids": [ 56 ], "parent_row_id": 56, "packet_id": 18, "LID_evidence": [ { "region_id": 44, "probabilities": { "ja": 0.3560518695782601, "zh": 0.005039430592079681, "en": 0.37702566033546275 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 44 ], "language_confidence": 0.3560518695782601, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 332.9, "packet_end": 361.8, "raw_text": "スライドを使用する際は両腕を前に伸ばし上体を起こして滑ります", "selected_text": "スライドを使用する際は両腕を前に伸ばし上体を起こして滑ります", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 56 ], "packet_id": 18, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "スライドを使用する際は両腕を前に伸ばし上体を起こして滑ります", "after": "スライドを使用する際は両腕を前に伸ばし上体を起こして滑ります", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard safety announcement phrase matches raw ASR and context perfectly." } ], "language": "ja", "packet_ids": [ 18 ], "index": 56, "source_text": "スライドを使用する際は両腕を前に伸ばし上体を起こして滑ります", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard safety announcement phrase matches raw ASR and context perfectly." }, { "start": 348.7, "end": 352.64, "text": "援助する場合はスライドの横に立ち体を引き起こしてください", "raw_asr_text": "援助する場合はスライドの横に立ち体を引き起こしてください", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 18, "words": [ { "word": "援", "start": 348.7, "end": 348.97999999999996, "probability": 0.98486328125 }, { "word": "助", "start": 348.97999999999996, "end": 349.09999999999997, "probability": 0.99853515625 }, { "word": "する", "start": 349.09999999999997, "end": 349.29999999999995, "probability": 0.99951171875 }, { "word": "場", "start": 349.29999999999995, "end": 349.5, "probability": 0.99951171875 }, { "word": "合", "start": 349.5, "end": 349.59999999999997, "probability": 1.0 }, { "word": "は", "start": 349.59999999999997, "end": 349.78, "probability": 0.99951171875 }, { "word": "ス", "start": 349.78, "end": 350.03999999999996, "probability": 0.99951171875 }, { "word": "ライ", "start": 350.03999999999996, "end": 350.14, "probability": 1.0 }, { "word": "ド", "start": 350.14, "end": 350.26, "probability": 1.0 }, { "word": "の", "start": 350.26, "end": 350.38, "probability": 0.99951171875 }, { "word": "横", "start": 350.38, "end": 350.59999999999997, "probability": 1.0 }, { "word": "に", "start": 350.59999999999997, "end": 350.7, "probability": 1.0 }, { "word": "立", "start": 350.7, "end": 350.82, "probability": 1.0 }, { "word": "ち", "start": 350.82, "end": 350.97999999999996, "probability": 1.0 }, { "word": "体", "start": 350.97999999999996, "end": 351.41999999999996, "probability": 0.94482421875 }, { "word": "を", "start": 351.41999999999996, "end": 351.65999999999997, "probability": 1.0 }, { "word": "引", "start": 351.65999999999997, "end": 351.78, "probability": 0.99951171875 }, { "word": "き", "start": 351.78, "end": 351.9, "probability": 1.0 }, { "word": "起", "start": 351.9, "end": 351.97999999999996, "probability": 1.0 }, { "word": "こ", "start": 351.97999999999996, "end": 352.08, "probability": 0.99951171875 }, { "word": "して", "start": 352.08, "end": 352.21999999999997, "probability": 1.0 }, { "word": "ください", "start": 352.21999999999997, "end": 352.64, "probability": 1.0 } ], "asr_metrics": { "avg_logprob": -0.1494094887558295, "compression_ratio": 1.4492753623188406, "no_speech_prob": 1.1984709231971369e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "手荷物は一切持たずハイヒールは脱いでくださいスライドを使用する際は両腕を前に伸ばし上体を起こして滑ります援助する場合はスライドの横に立ち体を引き起こしてください立ち上がりすぐに飛行機から離れてください行きましょう。" }, { "language": "en", "text": "Leave all baggage behind and remove high-heeled shoes. Put both arms out, sit up straight and slide down. Help the others up to their feet. Move away from the aircraft immediately. immediately." } ], "id": 57, "review_annotation": "", "row_id": 57, "bounded_row_id": 57, "parent_row_ids": [ 57 ], "parent_row_id": 57, "packet_id": 18, "LID_evidence": [ { "region_id": 44, "probabilities": { "ja": 0.3560518695782601, "zh": 0.005039430592079681, "en": 0.37702566033546275 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 44 ], "language_confidence": 0.3560518695782601, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 332.9, "packet_end": 361.8, "raw_text": "援助する場合はスライドの横に立ち体を引き起こしてください", "selected_text": "援助する場合はスライドの横に立ち体を引き起こしてください", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 57 ], "packet_id": 18, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "援助する場合はスライドの横に立ち体を引き起こしてください", "after": "援助する場合はスライドの横に立ち体を引き起こしてください", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard safety announcement phrase matches raw ASR and context perfectly." } ], "language": "ja", "packet_ids": [ 18 ], "index": 57, "source_text": "援助する場合はスライドの横に立ち体を引き起こしてください", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard safety announcement phrase matches raw ASR and context perfectly." }, { "start": 355.35999999999996, "end": 357.88, "text": "立ち上がりすぐに飛行機から離れてください", "raw_asr_text": "立ち上がりすぐに飛行機から離れてください", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 18, "words": [ { "word": "立", "start": 355.35999999999996, "end": 355.64, "probability": 0.919921875 }, { "word": "ち", "start": 355.64, "end": 355.76, "probability": 1.0 }, { "word": "上", "start": 355.76, "end": 355.88, "probability": 1.0 }, { "word": "が", "start": 355.88, "end": 356.02, "probability": 0.9970703125 }, { "word": "り", "start": 356.02, "end": 356.09999999999997, "probability": 0.99609375 }, { "word": "す", "start": 356.09999999999997, "end": 356.29999999999995, "probability": 0.998046875 }, { "word": "ぐ", "start": 356.29999999999995, "end": 356.41999999999996, "probability": 1.0 }, { "word": "に", "start": 356.41999999999996, "end": 356.53999999999996, "probability": 0.9990234375 }, { "word": "飛", "start": 356.53999999999996, "end": 356.64, "probability": 0.99951171875 }, { "word": "行", "start": 356.64, "end": 356.78, "probability": 0.99951171875 }, { "word": "機", "start": 356.78, "end": 356.9, "probability": 0.99951171875 }, { "word": "から", "start": 356.9, "end": 357.08, "probability": 1.0 }, { "word": "離", "start": 357.08, "end": 357.26, "probability": 1.0 }, { "word": "れて", "start": 357.26, "end": 357.46, "probability": 0.99951171875 }, { "word": "ください", "start": 357.46, "end": 357.88, "probability": 1.0 } ], "asr_metrics": { "avg_logprob": -0.1494094887558295, "compression_ratio": 1.4492753623188406, "no_speech_prob": 1.1984709231971369e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "手荷物は一切持たずハイヒールは脱いでくださいスライドを使用する際は両腕を前に伸ばし上体を起こして滑ります援助する場合はスライドの横に立ち体を引き起こしてください立ち上がりすぐに飛行機から離れてください行きましょう。" }, { "language": "en", "text": "Leave all baggage behind and remove high-heeled shoes. Put both arms out, sit up straight and slide down. Help the others up to their feet. Move away from the aircraft immediately. immediately." } ], "id": 58, "review_annotation": "", "row_id": 58, "bounded_row_id": 58, "parent_row_ids": [ 58 ], "parent_row_id": 58, "packet_id": 18, "LID_evidence": [ { "region_id": 45, "probabilities": { "ja": 0.22210561491422898, "zh": 0.003976678844871155, "en": 0.3891131089825649 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 45 ], "language_confidence": 0.22210561491422898, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 332.9, "packet_end": 361.8, "raw_text": "立ち上がりすぐに飛行機から離れてください", "selected_text": "立ち上がりすぐに飛行機から離れてください", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 58 ], "packet_id": 18, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "立ち上がりすぐに飛行機から離れてください", "after": "立ち上がりすぐに飛行機から離れてください", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard safety announcement phrase matches raw ASR and context perfectly." } ], "language": "ja", "packet_ids": [ 18 ], "index": 58, "source_text": "立ち上がりすぐに飛行機から離れてください", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard safety announcement phrase matches raw ASR and context perfectly." }, { "start": 357.88, "end": 358.78, "text": "行きましょう。", "raw_asr_text": "行きましょう。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 18, "words": [ { "word": "行", "start": 357.88, "end": 358.56, "probability": 0.053741455078125 }, { "word": "き", "start": 358.56, "end": 358.62, "probability": 0.2176513671875 }, { "word": "ましょう。", "start": 358.62, "end": 358.78, "probability": 0.262451171875 } ], "asr_metrics": { "avg_logprob": -1.3446264266967773, "compression_ratio": 0.7, "no_speech_prob": 1.1792254672737812e-10 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "手荷物は一切持たずハイヒールは脱いでくださいスライドを使用する際は両腕を前に伸ばし上体を起こして滑ります援助する場合はスライドの横に立ち体を引き起こしてください立ち上がりすぐに飛行機から離れてください行きましょう。" }, { "language": "en", "text": "Leave all baggage behind and remove high-heeled shoes. Put both arms out, sit up straight and slide down. Help the others up to their feet. Move away from the aircraft immediately. immediately." } ], "id": 59, "review_annotation": "待確認", "row_id": 59, "bounded_row_id": 59, "parent_row_ids": [ 59 ], "parent_row_id": 59, "packet_id": 18, "LID_evidence": [ { "region_id": 45, "probabilities": { "ja": 0.22210561491422898, "zh": 0.003976678844871155, "en": 0.3891131089825649 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 45 ], "language_confidence": 0.22210561491422898, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 332.9, "packet_end": 361.8, "raw_text": "行きましょう。", "selected_text": "行きましょう。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 59 ], "packet_id": 18, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "行きましょう。", "after": "行きましょう。", "classification": "BEST_SUPPORTED_WITH_REVIEW", "reason": "ASR output is nonsense; context implies 'Let's go' or 'Proceed', but exact wording is uncertain." } ], "language": "ja", "packet_ids": [ 18 ], "index": 59, "independent_ASR": { "engine": "kotoba-whisper", "model_path": "/Users/sagawa/AI_WORK/cache/huggingface/hub/models--kaiinui--kotoba-whisper-v2.0-mlx/snapshots/1a5ac6e52394898ef8996212e849d964e0dfb9ae", "language": "ja", "audio_interval": [ 332.9, 361.8 ], "target_interval": [ 333.7, 361.0 ], "text": "手荷物は一切持たずハイヒールは脱いでくださいスライドを使用する際は両腕を前に伸ばし上体を起こして滑ります救命をする場合はスライドの横に立ち体を引き起こしてください立ち上がりすぐに飛行機から離れてくださいもう行方からイメージュー", "words": [ { "word": "手", "start": 333.7, "end": 334.02, "probability": 0.96875 }, { "word": "荷", "start": 334.02, "end": 334.17999999999995, "probability": 0.9921875 }, { "word": "物", "start": 334.17999999999995, "end": 334.28, "probability": 0.9921875 }, { "word": "は", "start": 334.28, "end": 334.47999999999996, "probability": 0.99951171875 }, { "word": "一", "start": 334.47999999999996, "end": 334.65999999999997, "probability": 0.9990234375 }, { "word": "切", "start": 334.65999999999997, "end": 334.79999999999995, "probability": 0.9990234375 }, { "word": "持", "start": 334.79999999999995, "end": 334.97999999999996, "probability": 0.9130859375 }, { "word": "た", "start": 334.97999999999996, "end": 335.02, "probability": 1.0 }, { "word": "ず", "start": 335.02, "end": 335.21999999999997, "probability": 1.0 }, { "word": "ハ", "start": 335.21999999999997, "end": 335.5, "probability": 0.8701171875 }, { "word": "イ", "start": 335.5, "end": 335.5, "probability": 1.0 }, { "word": "ヒ", "start": 335.5, "end": 335.62, "probability": 0.99951171875 }, { "word": "ール", "start": 335.62, "end": 335.76, "probability": 0.9990234375 }, { "word": "は", "start": 335.76, "end": 335.91999999999996, "probability": 0.99609375 }, { "word": "脱", "start": 335.91999999999996, "end": 336.06, "probability": 0.86279296875 }, { "word": "い", "start": 336.06, "end": 336.08, "probability": 1.0 }, { "word": "で", "start": 336.08, "end": 336.17999999999995, "probability": 0.99951171875 }, { "word": "ください", "start": 336.17999999999995, "end": 336.58, "probability": 0.998046875 }, { "word": "ス", "start": 336.58, "end": 337.03999999999996, "probability": 0.51513671875 }, { "word": "ライ", "start": 337.03999999999996, "end": 337.03999999999996, "probability": 0.99951171875 }, { "word": "ド", "start": 337.03999999999996, "end": 341.02, "probability": 0.9990234375 }, { "word": "を", "start": 341.02, "end": 341.15999999999997, "probability": 0.9990234375 }, { "word": "使", "start": 341.15999999999997, "end": 341.26, "probability": 0.9921875 }, { "word": "用", "start": 341.26, "end": 341.26, "probability": 0.99951171875 }, { "word": "する", "start": 341.26, "end": 341.56, "probability": 0.99951171875 }, { "word": "際", "start": 341.56, "end": 341.9, "probability": 1.0 }, { "word": "は", "start": 341.9, "end": 342.02, "probability": 0.99951171875 }, { "word": "両", "start": 342.02, "end": 342.44, "probability": 0.998046875 }, { "word": "腕", "start": 342.44, "end": 342.53999999999996, "probability": 0.9970703125 }, { "word": "を", "start": 342.53999999999996, "end": 342.7, "probability": 0.9970703125 }, { "word": "前", "start": 342.7, "end": 342.88, "probability": 0.9990234375 }, { "word": "に", "start": 342.88, "end": 342.97999999999996, "probability": 1.0 }, { "word": "伸", "start": 342.97999999999996, "end": 343.14, "probability": 0.998046875 }, { "word": "ば", "start": 343.14, "end": 343.21999999999997, "probability": 1.0 }, { "word": "し", "start": 343.21999999999997, "end": 343.47999999999996, "probability": 0.99951171875 }, { "word": "上", "start": 343.47999999999996, "end": 343.71999999999997, "probability": 0.97607421875 }, { "word": "体", "start": 343.71999999999997, "end": 343.84, "probability": 0.99853515625 }, { "word": "を", "start": 343.84, "end": 344.0, "probability": 1.0 }, { "word": "起", "start": 344.0, "end": 344.17999999999995, "probability": 0.99951171875 }, { "word": "こ", "start": 344.17999999999995, "end": 344.17999999999995, "probability": 1.0 }, { "word": "して", "start": 344.17999999999995, "end": 344.38, "probability": 1.0 }, { "word": "滑", "start": 344.38, "end": 344.59999999999997, "probability": 0.916015625 }, { "word": "ります", "start": 344.59999999999997, "end": 344.88, "probability": 0.970703125 }, { "word": "救", "start": 345.41999999999996, "end": 345.73999999999995, "probability": 0.22412109375 }, { "word": "命", "start": 345.73999999999995, "end": 345.84, "probability": 0.99658203125 }, { "word": "を", "start": 345.84, "end": 346.34, "probability": 0.59765625 }, { "word": "する", "start": 346.34, "end": 346.38, "probability": 0.056549072265625 }, { "word": "場", "start": 346.38, "end": 346.4, "probability": 0.99951171875 }, { "word": "合", "start": 346.4, "end": 349.79999999999995, "probability": 1.0 }, { "word": "は", "start": 349.79999999999995, "end": 349.79999999999995, "probability": 0.9990234375 }, { "word": "ス", "start": 349.79999999999995, "end": 350.03999999999996, "probability": 0.99169921875 }, { "word": "ライ", "start": 350.03999999999996, "end": 350.17999999999995, "probability": 0.99658203125 }, { "word": "ド", "start": 350.17999999999995, "end": 350.28, "probability": 1.0 }, { "word": "の", "start": 350.28, "end": 350.4, "probability": 1.0 }, { "word": "横", "start": 350.4, "end": 350.58, "probability": 1.0 }, { "word": "に", "start": 350.58, "end": 350.7, "probability": 1.0 }, { "word": "立", "start": 350.7, "end": 350.85999999999996, "probability": 1.0 }, { "word": "ち", "start": 350.85999999999996, "end": 351.06, "probability": 1.0 }, { "word": "体", "start": 351.06, "end": 351.56, "probability": 0.97412109375 }, { "word": "を", "start": 351.56, "end": 351.65999999999997, "probability": 1.0 }, { "word": "引", "start": 351.65999999999997, "end": 351.85999999999996, "probability": 1.0 }, { "word": "き", "start": 351.85999999999996, "end": 351.85999999999996, "probability": 1.0 }, { "word": "起", "start": 351.85999999999996, "end": 352.08, "probability": 0.99951171875 }, { "word": "こ", "start": 352.08, "end": 352.09999999999997, "probability": 0.99951171875 }, { "word": "して", "start": 352.09999999999997, "end": 352.26, "probability": 1.0 }, { "word": "ください", "start": 352.26, "end": 352.73999999999995, "probability": 0.99951171875 }, { "word": "立", "start": 352.73999999999995, "end": 353.17999999999995, "probability": 0.7119140625 }, { "word": "ち", "start": 353.17999999999995, "end": 353.23999999999995, "probability": 1.0 }, { "word": "上", "start": 353.23999999999995, "end": 353.23999999999995, "probability": 1.0 }, { "word": "が", "start": 353.23999999999995, "end": 353.23999999999995, "probability": 0.99658203125 }, { "word": "り", "start": 353.23999999999995, "end": 356.09999999999997, "probability": 1.0 }, { "word": "す", "start": 356.09999999999997, "end": 356.34, "probability": 0.9990234375 }, { "word": "ぐ", "start": 356.34, "end": 356.41999999999996, "probability": 1.0 }, { "word": "に", "start": 356.41999999999996, "end": 356.58, "probability": 1.0 }, { "word": "飛", "start": 356.58, "end": 356.79999999999995, "probability": 0.99951171875 }, { "word": "行", "start": 356.79999999999995, "end": 356.84, "probability": 0.9990234375 }, { "word": "機", "start": 356.84, "end": 356.94, "probability": 0.9990234375 }, { "word": "から", "start": 356.94, "end": 357.08, "probability": 1.0 }, { "word": "離", "start": 357.08, "end": 357.34, "probability": 1.0 }, { "word": "れて", "start": 357.34, "end": 357.47999999999996, "probability": 1.0 }, { "word": "ください", "start": 357.47999999999996, "end": 357.84, "probability": 0.99951171875 }, { "word": "もう", "start": 357.84, "end": 358.52, "probability": 0.1195068359375 }, { "word": " 行", "start": 358.94, "end": 358.94, "probability": 0.0977783203125 }, { "word": "方", "start": 358.94, "end": 358.94, "probability": 0.53173828125 }, { "word": "から", "start": 358.94, "end": 359.08, "probability": 0.6953125 }, { "word": " イ", "start": 359.97999999999996, "end": 360.14, "probability": 0.1488037109375 }, { "word": "メ", "start": 360.14, "end": 360.26, "probability": 0.442626953125 }, { "word": "ージ", "start": 360.26, "end": 360.29999999999995, "probability": 0.953125 }, { "word": "ュ", "start": 360.29999999999995, "end": 360.53999999999996, "probability": 0.2039794921875 }, { "word": "ー", "start": 360.53999999999996, "end": 360.53999999999996, "probability": 0.296875 } ], "raw_result": { "text": "手荷物は一切持たずハイヒールは脱いでくださいスライドを使用する際は両腕を前に伸ばし上体を起こして滑ります救命をする場合はスライドの横に立ち体を引き起こしてください立ち上がりすぐに飛行機から離れてくださいもう 行方から イメージューリー", "segments": [ { "id": 0, "seek": 0, "start": 0.7999999999999998, "end": 3.68, "text": "手荷物は一切持たずハイヒールは脱いでください", "tokens": [ 50365, 11389, 31409, 115, 23516, 3065, 2257, 23632, 17694, 3368, 18216, 15927, 8040, 46995, 35307, 3065, 27067, 109, 1764, 2474, 25079, 50549 ], "temperature": 0.0, "avg_logprob": -0.16887556337842755, "compression_ratio": 1.4497607655502391, "no_speech_prob": 0.12614929676055908, "words": [ { "word": "手", "start": 0.7999999999999998, "end": 1.12, "probability": 0.96875 }, { "word": "荷", "start": 1.12, "end": 1.28, "probability": 0.9921875 }, { "word": "物", "start": 1.28, "end": 1.38, "probability": 0.9921875 }, { "word": "は", "start": 1.38, "end": 1.58, "probability": 0.99951171875 }, { "word": "一", "start": 1.58, "end": 1.76, "probability": 0.9990234375 }, { "word": "切", "start": 1.76, "end": 1.9, "probability": 0.9990234375 }, { "word": "持", "start": 1.9, "end": 2.08, "probability": 0.9130859375 }, { "word": "た", "start": 2.08, "end": 2.12, "probability": 1.0 }, { "word": "ず", "start": 2.12, "end": 2.32, "probability": 1.0 }, { "word": "ハ", "start": 2.32, "end": 2.6, "probability": 0.8701171875 }, { "word": "イ", "start": 2.6, "end": 2.6, "probability": 1.0 }, { "word": "ヒ", "start": 2.6, "end": 2.72, "probability": 0.99951171875 }, { "word": "ール", "start": 2.72, "end": 2.86, "probability": 0.9990234375 }, { "word": "は", "start": 2.86, "end": 3.02, "probability": 0.99609375 }, { "word": "脱", "start": 3.02, "end": 3.16, "probability": 0.86279296875 }, { "word": "い", "start": 3.16, "end": 3.18, "probability": 1.0 }, { "word": "で", "start": 3.18, "end": 3.28, "probability": 0.99951171875 }, { "word": "ください", "start": 3.28, "end": 3.68, "probability": 0.998046875 } ] }, { "id": 1, "seek": 0, "start": 3.68, "end": 9.12, "text": "スライドを使用する際は", "tokens": [ 50549, 9550, 40230, 11195, 5998, 22982, 9254, 22570, 34837, 3065, 50822 ], "temperature": 0.0, "avg_logprob": -0.16887556337842755, "compression_ratio": 1.4497607655502391, "no_speech_prob": 0.12614929676055908, "words": [ { "word": "ス", "start": 3.68, "end": 4.14, "probability": 0.51513671875 }, { "word": "ライ", "start": 4.14, "end": 4.14, "probability": 0.99951171875 }, { "word": "ド", "start": 4.14, "end": 8.12, "probability": 0.9990234375 }, { "word": "を", "start": 8.12, "end": 8.26, "probability": 0.9990234375 }, { "word": "使", "start": 8.26, "end": 8.36, "probability": 0.9921875 }, { "word": "用", "start": 8.36, "end": 8.36, "probability": 0.99951171875 }, { "word": "する", "start": 8.36, "end": 8.66, "probability": 0.99951171875 }, { "word": "際", "start": 8.66, "end": 9.0, "probability": 1.0 }, { "word": "は", "start": 9.0, "end": 9.12, "probability": 0.99951171875 } ] }, { "id": 2, "seek": 0, "start": 9.12, "end": 11.98, "text": "両腕を前に伸ばし上体を起こして滑ります", "tokens": [ 50822, 940, 94, 164, 12831, 5998, 8945, 4108, 7384, 116, 13349, 2849, 5708, 29485, 5998, 9147, 3503, 8822, 15868, 239, 19420, 50966 ], "temperature": 0.0, "avg_logprob": -0.16887556337842755, "compression_ratio": 1.4497607655502391, "no_speech_prob": 0.12614929676055908, "words": [ { "word": "両", "start": 9.12, "end": 9.54, "probability": 0.998046875 }, { "word": "腕", "start": 9.54, "end": 9.64, "probability": 0.9970703125 }, { "word": "を", "start": 9.64, "end": 9.8, "probability": 0.9970703125 }, { "word": "前", "start": 9.8, "end": 9.98, "probability": 0.9990234375 }, { "word": "に", "start": 9.98, "end": 10.08, "probability": 1.0 }, { "word": "伸", "start": 10.08, "end": 10.24, "probability": 0.998046875 }, { "word": "ば", "start": 10.24, "end": 10.32, "probability": 1.0 }, { "word": "し", "start": 10.32, "end": 10.58, "probability": 0.99951171875 }, { "word": "上", "start": 10.58, "end": 10.82, "probability": 0.97607421875 }, { "word": "体", "start": 10.82, "end": 10.94, "probability": 0.99853515625 }, { "word": "を", "start": 10.94, "end": 11.1, "probability": 1.0 }, { "word": "起", "start": 11.1, "end": 11.28, "probability": 0.99951171875 }, { "word": "こ", "start": 11.28, "end": 11.28, "probability": 1.0 }, { "word": "して", "start": 11.28, "end": 11.48, "probability": 1.0 }, { "word": "滑", "start": 11.48, "end": 11.7, "probability": 0.916015625 }, { "word": "ります", "start": 11.7, "end": 11.98, "probability": 0.970703125 } ] }, { "id": 3, "seek": 0, "start": 12.52, "end": 18.16, "text": "救命をする場合はスライドの横に立ち", "tokens": [ 50966, 39716, 25236, 5998, 22570, 21446, 14245, 3065, 9550, 40230, 11195, 2972, 5117, 103, 4108, 24409, 6574, 51272 ], "temperature": 0.0, "avg_logprob": -0.16887556337842755, "compression_ratio": 1.4497607655502391, "no_speech_prob": 0.12614929676055908, "words": [ { "word": "救", "start": 12.52, "end": 12.84, "probability": 0.22412109375 }, { "word": "命", "start": 12.84, "end": 12.94, "probability": 0.99658203125 }, { "word": "を", "start": 12.94, "end": 13.44, "probability": 0.59765625 }, { "word": "する", "start": 13.44, "end": 13.48, "probability": 0.056549072265625 }, { "word": "場", "start": 13.48, "end": 13.5, "probability": 0.99951171875 }, { "word": "合", "start": 13.5, "end": 16.9, "probability": 1.0 }, { "word": "は", "start": 16.9, "end": 16.9, "probability": 0.9990234375 }, { "word": "ス", "start": 16.9, "end": 17.14, "probability": 0.99169921875 }, { "word": "ライ", "start": 17.14, "end": 17.28, "probability": 0.99658203125 }, { "word": "ド", "start": 17.28, "end": 17.38, "probability": 1.0 }, { "word": "の", "start": 17.38, "end": 17.5, "probability": 1.0 }, { "word": "横", "start": 17.5, "end": 17.68, "probability": 1.0 }, { "word": "に", "start": 17.68, "end": 17.8, "probability": 1.0 }, { "word": "立", "start": 17.8, "end": 17.96, "probability": 1.0 }, { "word": "ち", "start": 17.96, "end": 18.16, "probability": 1.0 } ] }, { "id": 4, "seek": 0, "start": 18.16, "end": 19.84, "text": "体を引き起こしてください", "tokens": [ 51272, 29485, 5998, 41301, 7016, 9147, 3503, 8822, 25079, 51352 ], "temperature": 0.0, "avg_logprob": -0.16887556337842755, "compression_ratio": 1.4497607655502391, "no_speech_prob": 0.12614929676055908, "words": [ { "word": "体", "start": 18.16, "end": 18.66, "probability": 0.97412109375 }, { "word": "を", "start": 18.66, "end": 18.76, "probability": 1.0 }, { "word": "引", "start": 18.76, "end": 18.96, "probability": 1.0 }, { "word": "き", "start": 18.96, "end": 18.96, "probability": 1.0 }, { "word": "起", "start": 18.96, "end": 19.18, "probability": 0.99951171875 }, { "word": "こ", "start": 19.18, "end": 19.2, "probability": 0.99951171875 }, { "word": "して", "start": 19.2, "end": 19.36, "probability": 1.0 }, { "word": "ください", "start": 19.36, "end": 19.84, "probability": 0.99951171875 } ] }, { "id": 5, "seek": 0, "start": 19.84, "end": 24.94, "text": "立ち上がりすぐに飛行機から離れてください", "tokens": [ 51352, 24409, 6574, 5708, 5142, 5095, 2659, 35849, 4108, 34629, 8082, 17543, 11046, 32662, 26638, 25079, 51612 ], "temperature": 0.0, "avg_logprob": -0.16887556337842755, "compression_ratio": 1.4497607655502391, "no_speech_prob": 0.12614929676055908, "words": [ { "word": "立", "start": 19.84, "end": 20.28, "probability": 0.7119140625 }, { "word": "ち", "start": 20.28, "end": 20.34, "probability": 1.0 }, { "word": "上", "start": 20.34, "end": 20.34, "probability": 1.0 }, { "word": "が", "start": 20.34, "end": 20.34, "probability": 0.99658203125 }, { "word": "り", "start": 20.34, "end": 23.2, "probability": 1.0 }, { "word": "す", "start": 23.2, "end": 23.44, "probability": 0.9990234375 }, { "word": "ぐ", "start": 23.44, "end": 23.52, "probability": 1.0 }, { "word": "に", "start": 23.52, "end": 23.68, "probability": 1.0 }, { "word": "飛", "start": 23.68, "end": 23.9, "probability": 0.99951171875 }, { "word": "行", "start": 23.9, "end": 23.94, "probability": 0.9990234375 }, { "word": "機", "start": 23.94, "end": 24.04, "probability": 0.9990234375 }, { "word": "から", "start": 24.04, "end": 24.18, "probability": 1.0 }, { "word": "離", "start": 24.18, "end": 24.44, "probability": 1.0 }, { "word": "れて", "start": 24.44, "end": 24.58, "probability": 1.0 }, { "word": "ください", "start": 24.58, "end": 24.94, "probability": 0.99951171875 } ] }, { "id": 6, "seek": 2494, "start": 24.94, "end": 28.88, "text": "もう 行方から イメージューリー", "tokens": [ 50365, 16324, 220, 8082, 9249, 11046, 220, 8040, 21647, 44143, 26167, 3384, 12376, 3384, 50565 ], "temperature": 0.0, "avg_logprob": -0.9573916792869568, "compression_ratio": 0.8627450980392157, "no_speech_prob": 0.09706871956586838, "words": [ { "word": "もう", "start": 24.94, "end": 25.62, "probability": 0.1195068359375 }, { "word": " 行", "start": 26.04, "end": 26.04, "probability": 0.0977783203125 }, { "word": "方", "start": 26.04, "end": 26.04, "probability": 0.53173828125 }, { "word": "から", "start": 26.04, "end": 26.18, "probability": 0.6953125 }, { "word": " イ", "start": 27.08, "end": 27.24, "probability": 0.1488037109375 }, { "word": "メ", "start": 27.24, "end": 27.36, "probability": 0.442626953125 }, { "word": "ージ", "start": 27.36, "end": 27.4, "probability": 0.953125 }, { "word": "ュ", "start": 27.4, "end": 27.64, "probability": 0.2039794921875 }, { "word": "ー", "start": 27.64, "end": 27.64, "probability": 0.296875 }, { "word": "リ", "start": 27.64, "end": 28.84, "probability": 0.32080078125 }, { "word": "ー", "start": 28.84, "end": 28.88, "probability": 0.7138671875 } ] } ], "language": "ja", "duration": 28.9 }, "status": "COMPLETE", "rejected_alternative": { "text": "手荷物は一切持たずハイヒールは脱いでください。 Leave all baggage behind and remove high-heeled shoes. スライドを使用する際は両腕を前に伸ばし上体を起こして滑ります。 Put both arms out, sit up straight, and slide down. 救助する場合はスライドの横に立ち体を引き起こしてください。 Help the others up to their feet. 立ち上がりすぐに飛行機から離れてください。 Move away from the aircraft immediately. 救命", "chunks": [ { "text": "手荷物は一切持たずハイヒールは脱いでください。 Leave all baggage behind and remove high-heeled shoes. スライドを使用する際は両腕を前に伸ばし上体を起こして滑ります。 Put both arms out, sit up straight, and slide down. 救助する場合はスライドの横に立ち体を引き起こしてください。 Help the others up to their feet. 立ち上がりすぐに飛行機から離れてください。 Move away from the aircraft immediately. 救命", "start": 0.0, "end": 28.9, "chunk_index": 0, "language": "Japanese", "finish_reason": "eos", "truncated": false, "generated_tokens": 113, "max_new_tokens": 379 } ], "detected_language": "Japanese", "segments": [ { "start": 332.9, "end": 361.79999999999995, "text": "手荷物は一切持たずハイヒールは脱いでください。 Leave all baggage behind and remove high-heeled shoes. スライドを使用する際は両腕を前に伸ばし上体を起こして滑ります。 Put both arms out, sit up straight, and slide down. 救助する場合はスライドの横に立ち体を引き起こしてください。 Help the others up to their feet. 立ち上がりすぐに飛行機から離れてください。 Move away from the aircraft immediately. 救命" } ], "start": 332.9, "end": 361.8, "engine": "mlx-qwen3-asr", "language": "ja" } }, "source_text": "行きましょう。", "review_status": "BEST_SUPPORTED_WITH_REVIEW", "classification": "BEST_SUPPORTED_WITH_REVIEW", "confidence": 0.8, "correction_reason": "ASR output is nonsense; context implies 'Let's go' or 'Proceed', but exact wording is uncertain." }, { "start": 361.2, "end": 363.94, "text": "救命衣は座席の下または横にあります", "raw_asr_text": "救命衣は座席の下または横にあります", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 19, "words": [ { "word": "救", "start": 361.2, "end": 361.52, "probability": 0.77734375 }, { "word": "命", "start": 361.52, "end": 361.73999999999995, "probability": 0.9970703125 }, { "word": "衣", "start": 361.73999999999995, "end": 361.85999999999996, "probability": 0.81689453125 }, { "word": "は", "start": 361.85999999999996, "end": 362.0, "probability": 0.9970703125 }, { "word": "座", "start": 362.0, "end": 362.2, "probability": 0.9970703125 }, { "word": "席", "start": 362.2, "end": 362.4, "probability": 1.0 }, { "word": "の", "start": 362.4, "end": 362.52, "probability": 0.99951171875 }, { "word": "下", "start": 362.52, "end": 362.7, "probability": 1.0 }, { "word": "また", "start": 362.7, "end": 363.03999999999996, "probability": 0.9873046875 }, { "word": "は", "start": 363.03999999999996, "end": 363.17999999999995, "probability": 1.0 }, { "word": "横", "start": 363.17999999999995, "end": 363.41999999999996, "probability": 1.0 }, { "word": "に", "start": 363.41999999999996, "end": 363.53999999999996, "probability": 0.99951171875 }, { "word": "あります", "start": 363.53999999999996, "end": 363.94, "probability": 0.99951171875 } ], "asr_metrics": { "avg_logprob": -0.14851270595066984, "compression_ratio": 1.2692307692307692, "no_speech_prob": 1.5076386319923252e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "救命衣は座席の下または横にあります袋から取り出し背当て布を後ろにしてかぶります両側の金具を前の留め金にかけ左右に引いて締めます非常口" }, { "language": "en", "text": "A life vest is under or beside your seat. Slip the life vest over your head. Fasten the buckles and tighten the straps. I'm not sure how to do this." } ], "id": 60, "review_annotation": "", "row_id": 60, "bounded_row_id": 60, "parent_row_ids": [ 60 ], "parent_row_id": 60, "packet_id": 19, "LID_evidence": [ { "region_id": 46, "probabilities": { "ja": 0.2476968026276154, "zh": 0.004842027205540944, "en": 0.39529711733484874 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 46 ], "language_confidence": 0.2476968026276154, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 360.4, "packet_end": 380.90000000000003, "raw_text": "救命衣は座席の下または横にあります", "selected_text": "救命衣は座席の下または横にあります", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 60 ], "packet_id": 19, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "救命衣は座席の下または横にあります", "after": "救命衣は座席の下または横にあります", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard safety announcement phrasing matches context and ASR output perfectly." } ], "language": "ja", "packet_ids": [ 19 ], "index": 60, "source_text": "救命衣は座席の下または横にあります", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Standard safety announcement phrasing matches context and ASR output perfectly." }, { "start": 367.26, "end": 370.52, "text": "袋から取り出し背当て布を後ろにしてかぶります", "raw_asr_text": "袋から取り出し背当て布を後ろにしてかぶります", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 19, "words": [ { "word": "袋", "start": 367.26, "end": 367.58, "probability": 0.6533203125 }, { "word": "から", "start": 367.58, "end": 367.78, "probability": 1.0 }, { "word": "取", "start": 367.78, "end": 367.94, "probability": 1.0 }, { "word": "り", "start": 367.94, "end": 368.0, "probability": 1.0 }, { "word": "出", "start": 368.0, "end": 368.12, "probability": 1.0 }, { "word": "し", "start": 368.12, "end": 368.38, "probability": 1.0 }, { "word": "背", "start": 368.38, "end": 368.85999999999996, "probability": 0.55712890625 }, { "word": "当", "start": 368.85999999999996, "end": 368.97999999999996, "probability": 0.95166015625 }, { "word": "て", "start": 368.97999999999996, "end": 369.09999999999997, "probability": 0.982421875 }, { "word": "布", "start": 369.09999999999997, "end": 369.15999999999997, "probability": 0.9951171875 }, { "word": "を", "start": 369.15999999999997, "end": 369.35999999999996, "probability": 1.0 }, { "word": "後", "start": 369.35999999999996, "end": 369.53999999999996, "probability": 0.99951171875 }, { "word": "ろ", "start": 369.53999999999996, "end": 369.64, "probability": 0.99853515625 }, { "word": "に", "start": 369.64, "end": 369.76, "probability": 1.0 }, { "word": "して", "start": 369.76, "end": 369.9, "probability": 0.99951171875 }, { "word": "か", "start": 369.9, "end": 370.06, "probability": 0.822265625 }, { "word": "ぶ", "start": 370.06, "end": 370.14, "probability": 0.99853515625 }, { "word": "ります", "start": 370.14, "end": 370.52, "probability": 1.0 } ], "asr_metrics": { "avg_logprob": -0.14851270595066984, "compression_ratio": 1.2692307692307692, "no_speech_prob": 1.5076386319923252e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "救命衣は座席の下または横にあります袋から取り出し背当て布を後ろにしてかぶります両側の金具を前の留め金にかけ左右に引いて締めます非常口" }, { "language": "en", "text": "A life vest is under or beside your seat. Slip the life vest over your head. Fasten the buckles and tighten the straps. I'm not sure how to do this." } ], "id": 61, "review_annotation": "", "row_id": 61, "bounded_row_id": 61, "parent_row_ids": [ 61 ], "parent_row_id": 61, "packet_id": 19, "LID_evidence": [ { "region_id": 47, "probabilities": { "ja": 0.412451735237317, "zh": 0.004328784546735485, "en": 0.25062774775274604 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 47 ], "language_confidence": 0.412451735237317, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 360.4, "packet_end": 380.90000000000003, "raw_text": "袋から取り出し背当て布を後ろにしてかぶります", "selected_text": "袋から取り出し背当て布を後ろにしてかぶります", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 61 ], "packet_id": 19, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "袋から取り出し背当て布を後ろにしてかぶります", "after": "袋から取り出し背当て布を後ろにしてかぶります", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard life vest instruction; grammar and vocabulary are correct for the context." } ], "language": "ja", "packet_ids": [ 19 ], "index": 61, "source_text": "袋から取り出し背当て布を後ろにしてかぶります", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Standard life vest instruction; grammar and vocabulary are correct for the context." }, { "start": 373.59999999999997, "end": 377.46, "text": "両側の金具を前の留め金にかけ左右に引いて締めます", "raw_asr_text": "両側の金具を前の留め金にかけ左右に引いて締めます", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 19, "words": [ { "word": "両", "start": 373.59999999999997, "end": 373.91999999999996, "probability": 0.98046875 }, { "word": "側", "start": 373.91999999999996, "end": 374.14, "probability": 1.0 }, { "word": "の", "start": 374.14, "end": 374.26, "probability": 1.0 }, { "word": "金", "start": 374.26, "end": 374.41999999999996, "probability": 0.99853515625 }, { "word": "具", "start": 374.41999999999996, "end": 374.58, "probability": 1.0 }, { "word": "を", "start": 374.58, "end": 374.73999999999995, "probability": 1.0 }, { "word": "前", "start": 374.73999999999995, "end": 374.94, "probability": 0.9990234375 }, { "word": "の", "start": 374.94, "end": 375.08, "probability": 0.99951171875 }, { "word": "留", "start": 375.08, "end": 375.2, "probability": 0.90673828125 }, { "word": "め", "start": 375.2, "end": 375.34, "probability": 0.9873046875 }, { "word": "金", "start": 375.34, "end": 375.4, "probability": 0.99072265625 }, { "word": "に", "start": 375.4, "end": 375.59999999999997, "probability": 0.99951171875 }, { "word": "か", "start": 375.59999999999997, "end": 375.7, "probability": 0.96826171875 }, { "word": "け", "start": 375.7, "end": 375.88, "probability": 0.9990234375 }, { "word": "左右", "start": 375.88, "end": 376.35999999999996, "probability": 0.98828125 }, { "word": "に", "start": 376.35999999999996, "end": 376.56, "probability": 1.0 }, { "word": "引", "start": 376.56, "end": 376.67999999999995, "probability": 0.9853515625 }, { "word": "いて", "start": 376.67999999999995, "end": 376.88, "probability": 0.9990234375 }, { "word": "締", "start": 376.88, "end": 377.03999999999996, "probability": 0.98193359375 }, { "word": "め", "start": 377.03999999999996, "end": 377.12, "probability": 0.99951171875 }, { "word": "ます", "start": 377.12, "end": 377.46, "probability": 0.99951171875 } ], "asr_metrics": { "avg_logprob": -0.14851270595066984, "compression_ratio": 1.2692307692307692, "no_speech_prob": 1.5076386319923252e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "救命衣は座席の下または横にあります袋から取り出し背当て布を後ろにしてかぶります両側の金具を前の留め金にかけ左右に引いて締めます非常口" }, { "language": "en", "text": "A life vest is under or beside your seat. Slip the life vest over your head. Fasten the buckles and tighten the straps. I'm not sure how to do this." } ], "id": 62, "review_annotation": "", "row_id": 62, "bounded_row_id": 62, "parent_row_ids": [ 62 ], "parent_row_id": 62, "packet_id": 19, "LID_evidence": [ { "region_id": 48, "probabilities": { "ja": 0.28930798508041666, "zh": 0.0036745182971406557, "en": 0.3721223890289283 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 48 ], "language_confidence": 0.28930798508041666, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 360.4, "packet_end": 380.90000000000003, "raw_text": "両側の金具を前の留め金にかけ左右に引いて締めます", "selected_text": "両側の金具を前の留め金にかけ左右に引いて締めます", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 62 ], "packet_id": 19, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "両側の金具を前の留め金にかけ左右に引いて締めます", "after": "両側の金具を前の留め金にかけ左右に引いて締めます", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard life vest instruction; matches standard ANA safety script." } ], "language": "ja", "packet_ids": [ 19 ], "index": 62, "source_text": "両側の金具を前の留め金にかけ左右に引いて締めます", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Standard life vest instruction; matches standard ANA safety script." }, { "start": 380.23999999999995, "end": 384.64, "text": "非常口または翼の上で前の引き手を強く引いて膨らませてください", "raw_asr_text": "非常口または翼の上で前の引き手を強く引いて膨らませてください", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 20, "words": [ { "word": "非常", "start": 380.23999999999995, "end": 380.56, "probability": 0.87548828125 }, { "word": "口", "start": 380.56, "end": 380.84, "probability": 0.90234375 }, { "word": "また", "start": 380.84, "end": 381.14, "probability": 0.97509765625 }, { "word": "は", "start": 381.14, "end": 381.34, "probability": 0.9990234375 }, { "word": "翼", "start": 381.34, "end": 381.56, "probability": 0.98681640625 }, { "word": "の", "start": 381.56, "end": 381.71999999999997, "probability": 0.9990234375 }, { "word": "上", "start": 381.71999999999997, "end": 381.88, "probability": 1.0 }, { "word": "で", "start": 381.88, "end": 382.02, "probability": 1.0 }, { "word": "前", "start": 382.02, "end": 382.47999999999996, "probability": 0.91015625 }, { "word": "の", "start": 382.47999999999996, "end": 382.64, "probability": 0.99951171875 }, { "word": "引", "start": 382.64, "end": 382.73999999999995, "probability": 0.98583984375 }, { "word": "き", "start": 382.73999999999995, "end": 382.84, "probability": 0.99755859375 }, { "word": "手", "start": 382.84, "end": 382.9, "probability": 0.99951171875 }, { "word": "を", "start": 382.9, "end": 383.03999999999996, "probability": 0.99462890625 }, { "word": "強", "start": 383.03999999999996, "end": 383.2, "probability": 0.994140625 }, { "word": "く", "start": 383.2, "end": 383.4, "probability": 1.0 }, { "word": "引", "start": 383.4, "end": 383.47999999999996, "probability": 0.99560546875 }, { "word": "いて", "start": 383.47999999999996, "end": 383.7, "probability": 0.99951171875 }, { "word": "膨", "start": 383.7, "end": 383.94, "probability": 0.99609375 }, { "word": "ら", "start": 383.94, "end": 384.02, "probability": 0.99951171875 }, { "word": "ませ", "start": 384.02, "end": 384.17999999999995, "probability": 0.99853515625 }, { "word": "て", "start": 384.17999999999995, "end": 384.29999999999995, "probability": 0.99951171875 }, { "word": "ください", "start": 384.29999999999995, "end": 384.64, "probability": 0.998046875 } ], "asr_metrics": { "avg_logprob": -0.11740236918131511, "compression_ratio": 1.3703703703703705, "no_speech_prob": 3.623901526084161e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "非常口または翼の上で前の引き手を強く引いて膨らませてください膨らみが足りない場合は両側のゴム缶を吹いてください座席ポケットの安全の修理をご覧ください" }, { "language": "en", "text": "On the right side of the seat, the front side of the seat is a very strong one. If you need more air, blow into the tubes. Please look at the safety instructions card in your seat pocket." } ], "id": 63, "review_annotation": "", "row_id": 63, "bounded_row_id": 63, "parent_row_ids": [ 63 ], "parent_row_id": 63, "packet_id": 20, "LID_evidence": [ { "region_id": 49, "probabilities": { "ja": 0.2007864812425396, "zh": 0.004173079210262845, "en": 0.36917341764949796 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 49 ], "language_confidence": 0.2007864812425396, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 379.4, "packet_end": 404.0, "raw_text": "非常口または翼の上で前の引き手を強く引いて膨らませてください", "selected_text": "非常口または翼の上で前の引き手を強く引いて膨らませてください", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 63 ], "packet_id": 20, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "非常口または翼の上で前の引き手を強く引いて膨らませてください", "after": "非常口または翼の上で前の引き手を強く引いて膨らませてください", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard life vest inflation instruction; phrasing is grammatically correct." } ], "language": "ja", "packet_ids": [ 20 ], "index": 63, "source_text": "非常口または翼の上で前の引き手を強く引いて膨らませてください", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Standard life vest inflation instruction; phrasing is grammatically correct." }, { "start": 389.34, "end": 392.67999999999995, "text": "膨らみが足りない場合は両側のゴム管を吹いてください", "raw_asr_text": "膨らみが足りない場合は両側のゴム缶を吹いてください", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 20, "words": [], "asr_metrics": { "avg_logprob": -0.11740236918131511, "compression_ratio": 1.3703703703703705, "no_speech_prob": 3.623901526084161e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "非常口または翼の上で前の引き手を強く引いて膨らませてください膨らみが足りない場合は両側のゴム缶を吹いてください座席ポケットの安全の修理をご覧ください" }, { "language": "en", "text": "On the right side of the seat, the front side of the seat is a very strong one. If you need more air, blow into the tubes. Please look at the safety instructions card in your seat pocket." } ], "id": 64, "review_annotation": "", "row_id": 64, "bounded_row_id": 64, "parent_row_ids": [ 64 ], "parent_row_id": 64, "packet_id": 20, "LID_evidence": [ { "region_id": 49, "probabilities": { "ja": 0.2007864812425396, "zh": 0.004173079210262845, "en": 0.36917341764949796 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 49 ], "language_confidence": 0.2007864812425396, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 379.4, "packet_end": 404.0, "raw_text": "膨らみが足りない場合は両側のゴム缶を吹いてください", "selected_text": "膨らみが足りない場合は両側のゴム管を吹いてください", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 64 ], "packet_id": 20, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "膨らみが足りない場合は両側のゴム缶を吹いてください", "after": "膨らみが足りない場合は両側のゴム管を吹いてください", "classification": "STRONG_RESTORATION", "reason": "ASR 'ゴム缶' (can) is a phonetic error for 'ゴム管' (tube), standard life vest instruction." } ], "language": "ja", "packet_ids": [ 20 ], "index": 64, "independent_ASR": { "engine": "kotoba-whisper", "model_path": "/Users/sagawa/AI_WORK/cache/huggingface/hub/models--kaiinui--kotoba-whisper-v2.0-mlx/snapshots/1a5ac6e52394898ef8996212e849d964e0dfb9ae", "language": "ja", "audio_interval": [ 379.4, 404.0 ], "target_interval": [ 380.2, 403.2 ], "text": "非常口または翼の上で前の引き手を強く引いて膨らませてくださいフクラミが足りない場合は両側のゴム缶を吹いてください膨らみが足りない場合は両側のゴム缶をご覧くださいぜひご覧ください!!!!!", "words": [ { "word": "非常", "start": 380.2, "end": 380.56, "probability": 0.59423828125 }, { "word": "口", "start": 380.56, "end": 380.88, "probability": 0.9736328125 }, { "word": "また", "start": 380.88, "end": 381.17999999999995, "probability": 0.99609375 }, { "word": "は", "start": 381.17999999999995, "end": 381.34, "probability": 0.99755859375 }, { "word": "翼", "start": 381.34, "end": 381.5, "probability": 0.99169921875 }, { "word": "の", "start": 381.5, "end": 381.71999999999997, "probability": 0.99951171875 }, { "word": "上", "start": 381.71999999999997, "end": 381.9, "probability": 0.99951171875 }, { "word": "で", "start": 381.9, "end": 382.15999999999997, "probability": 0.99951171875 }, { "word": "前", "start": 382.15999999999997, "end": 382.52, "probability": 0.951171875 }, { "word": "の", "start": 382.52, "end": 382.7, "probability": 0.9990234375 }, { "word": "引", "start": 382.7, "end": 382.84, "probability": 0.98876953125 }, { "word": "き", "start": 382.84, "end": 382.84, "probability": 0.99951171875 }, { "word": "手", "start": 382.84, "end": 382.94, "probability": 0.99951171875 }, { "word": "を", "start": 382.94, "end": 383.03999999999996, "probability": 0.9990234375 }, { "word": "強", "start": 383.03999999999996, "end": 383.23999999999995, "probability": 0.99072265625 }, { "word": "く", "start": 383.23999999999995, "end": 383.4, "probability": 1.0 }, { "word": "引", "start": 383.4, "end": 383.56, "probability": 0.76904296875 }, { "word": "いて", "start": 383.56, "end": 383.7, "probability": 0.99951171875 }, { "word": "膨", "start": 383.7, "end": 383.97999999999996, "probability": 0.99072265625 }, { "word": "ら", "start": 383.97999999999996, "end": 384.03999999999996, "probability": 1.0 }, { "word": "ませ", "start": 384.03999999999996, "end": 384.26, "probability": 0.99951171875 }, { "word": "て", "start": 384.26, "end": 384.35999999999996, "probability": 0.99951171875 }, { "word": "ください", "start": 384.35999999999996, "end": 384.85999999999996, "probability": 0.998046875 }, { "word": "フ", "start": 384.85999999999996, "end": 385.14, "probability": 0.067138671875 }, { "word": "ク", "start": 385.14, "end": 385.21999999999997, "probability": 0.74072265625 }, { "word": "ラ", "start": 385.21999999999997, "end": 385.34, "probability": 0.9970703125 }, { "word": "ミ", "start": 385.34, "end": 385.52, "probability": 0.94482421875 }, { "word": "が", "start": 385.52, "end": 385.52, "probability": 0.99951171875 }, { "word": "足", "start": 385.52, "end": 385.91999999999996, "probability": 1.0 }, { "word": "り", "start": 385.91999999999996, "end": 386.09999999999997, "probability": 1.0 }, { "word": "ない", "start": 386.09999999999997, "end": 386.34, "probability": 0.99951171875 }, { "word": "場", "start": 386.34, "end": 386.62, "probability": 0.99951171875 }, { "word": "合", "start": 386.62, "end": 390.85999999999996, "probability": 1.0 }, { "word": "は", "start": 390.85999999999996, "end": 390.88, "probability": 0.99755859375 }, { "word": "両", "start": 390.88, "end": 391.26, "probability": 0.9990234375 }, { "word": "側", "start": 391.26, "end": 391.4, "probability": 1.0 }, { "word": "の", "start": 391.4, "end": 391.52, "probability": 1.0 }, { "word": "ゴ", "start": 391.52, "end": 391.7, "probability": 0.98828125 }, { "word": "ム", "start": 391.7, "end": 391.73999999999995, "probability": 1.0 }, { "word": "缶", "start": 391.73999999999995, "end": 391.9, "probability": 0.9814453125 }, { "word": "を", "start": 391.9, "end": 392.02, "probability": 1.0 }, { "word": "吹", "start": 392.02, "end": 392.14, "probability": 0.9111328125 }, { "word": "いて", "start": 392.14, "end": 392.23999999999995, "probability": 1.0 }, { "word": "ください", "start": 392.23999999999995, "end": 392.84, "probability": 0.99853515625 }, { "word": "膨", "start": 392.84, "end": 393.14, "probability": 0.505859375 }, { "word": "ら", "start": 393.14, "end": 393.26, "probability": 0.99755859375 }, { "word": "み", "start": 393.26, "end": 393.52, "probability": 0.9990234375 }, { "word": "が", "start": 393.52, "end": 393.62, "probability": 0.99951171875 }, { "word": "足", "start": 393.62, "end": 393.73999999999995, "probability": 1.0 }, { "word": "り", "start": 393.73999999999995, "end": 393.73999999999995, "probability": 1.0 }, { "word": "ない", "start": 393.73999999999995, "end": 393.84, "probability": 0.99951171875 }, { "word": "場", "start": 393.84, "end": 393.88, "probability": 1.0 }, { "word": "合", "start": 393.88, "end": 394.08, "probability": 1.0 }, { "word": "は", "start": 394.08, "end": 394.32, "probability": 0.998046875 }, { "word": "両", "start": 394.32, "end": 394.53999999999996, "probability": 0.9921875 }, { "word": "側", "start": 394.53999999999996, "end": 394.65999999999997, "probability": 1.0 }, { "word": "の", "start": 394.65999999999997, "end": 394.65999999999997, "probability": 0.99951171875 }, { "word": "ゴ", "start": 394.65999999999997, "end": 394.65999999999997, "probability": 0.99560546875 }, { "word": "ム", "start": 394.65999999999997, "end": 394.65999999999997, "probability": 1.0 }, { "word": "缶", "start": 394.65999999999997, "end": 394.65999999999997, "probability": 0.9921875 }, { "word": "を", "start": 394.65999999999997, "end": 395.08, "probability": 1.0 }, { "word": "ご", "start": 395.08, "end": 395.08, "probability": 0.050445556640625 }, { "word": "覧", "start": 395.08, "end": 395.21999999999997, "probability": 0.994140625 }, { "word": "ください", "start": 395.21999999999997, "end": 395.67999999999995, "probability": 0.99951171875 }, { "word": "ぜ", "start": 395.67999999999995, "end": 396.28, "probability": 0.0302886962890625 }, { "word": "ひ", "start": 396.28, "end": 396.46, "probability": 0.73583984375 }, { "word": "ご", "start": 396.46, "end": 396.65999999999997, "probability": 3.93986701965332e-05 }, { "word": "覧", "start": 396.65999999999997, "end": 396.65999999999997, "probability": 0.9873046875 }, { "word": "ください!!!!!", "start": 396.65999999999997, "end": 398.79999999999995, "probability": 0.99609375 } ], "raw_result": { "text": "非常口または翼の上で前の引き手を強く引いて膨らませてくださいフクラミが足りない場合は両側のゴム缶を吹いてください膨らみが足りない場合は両側のゴム缶をご覧くださいぜひご覧ください!!!", "segments": [ { "id": 0, "seek": 0, "start": 0.7999999999999996, "end": 5.46, "text": "非常口または翼の上で前の引き手を強く引いて膨らませてください", "tokens": [ 50365, 14392, 18144, 24712, 3065, 18573, 120, 2972, 5708, 2474, 8945, 2972, 41301, 7016, 11389, 5998, 22752, 6134, 41301, 18549, 39708, 101, 5154, 28742, 2996, 25079, 50633 ], "temperature": 0.0, "avg_logprob": NaN, "compression_ratio": 1.68125, "no_speech_prob": 0.14235487580299377, "words": [ { "word": "非常", "start": 0.7999999999999996, "end": 1.16, "probability": 0.59423828125 }, { "word": "口", "start": 1.16, "end": 1.48, "probability": 0.9736328125 }, { "word": "また", "start": 1.48, "end": 1.78, "probability": 0.99609375 }, { "word": "は", "start": 1.78, "end": 1.94, "probability": 0.99755859375 }, { "word": "翼", "start": 1.94, "end": 2.1, "probability": 0.99169921875 }, { "word": "の", "start": 2.1, "end": 2.32, "probability": 0.99951171875 }, { "word": "上", "start": 2.32, "end": 2.5, "probability": 0.99951171875 }, { "word": "で", "start": 2.5, "end": 2.76, "probability": 0.99951171875 }, { "word": "前", "start": 2.76, "end": 3.12, "probability": 0.951171875 }, { "word": "の", "start": 3.12, "end": 3.3, "probability": 0.9990234375 }, { "word": "引", "start": 3.3, "end": 3.44, "probability": 0.98876953125 }, { "word": "き", "start": 3.44, "end": 3.44, "probability": 0.99951171875 }, { "word": "手", "start": 3.44, "end": 3.54, "probability": 0.99951171875 }, { "word": "を", "start": 3.54, "end": 3.64, "probability": 0.9990234375 }, { "word": "強", "start": 3.64, "end": 3.84, "probability": 0.99072265625 }, { "word": "く", "start": 3.84, "end": 4.0, "probability": 1.0 }, { "word": "引", "start": 4.0, "end": 4.16, "probability": 0.76904296875 }, { "word": "いて", "start": 4.16, "end": 4.3, "probability": 0.99951171875 }, { "word": "膨", "start": 4.3, "end": 4.58, "probability": 0.99072265625 }, { "word": "ら", "start": 4.58, "end": 4.64, "probability": 1.0 }, { "word": "ませ", "start": 4.64, "end": 4.86, "probability": 0.99951171875 }, { "word": "て", "start": 4.86, "end": 4.96, "probability": 0.99951171875 }, { "word": "ください", "start": 4.96, "end": 5.46, "probability": 0.998046875 } ] }, { "id": 1, "seek": 0, "start": 5.46, "end": 13.44, "text": "フクラミが足りない場合は両側のゴム缶を吹いてください", "tokens": [ 50633, 17320, 10825, 11353, 23196, 5142, 37236, 5095, 9311, 21446, 14245, 3065, 940, 94, 7437, 112, 2972, 39780, 32026, 38109, 114, 5998, 2392, 117, 18549, 25079, 51030 ], "temperature": 0.0, "avg_logprob": NaN, "compression_ratio": 1.68125, "no_speech_prob": 0.14235487580299377, "words": [ { "word": "フ", "start": 5.46, "end": 5.74, "probability": 0.067138671875 }, { "word": "ク", "start": 5.74, "end": 5.82, "probability": 0.74072265625 }, { "word": "ラ", "start": 5.82, "end": 5.94, "probability": 0.9970703125 }, { "word": "ミ", "start": 5.94, "end": 6.12, "probability": 0.94482421875 }, { "word": "が", "start": 6.12, "end": 6.12, "probability": 0.99951171875 }, { "word": "足", "start": 6.12, "end": 6.52, "probability": 1.0 }, { "word": "り", "start": 6.52, "end": 6.7, "probability": 1.0 }, { "word": "ない", "start": 6.7, "end": 6.94, "probability": 0.99951171875 }, { "word": "場", "start": 6.94, "end": 7.22, "probability": 0.99951171875 }, { "word": "合", "start": 7.22, "end": 11.46, "probability": 1.0 }, { "word": "は", "start": 11.46, "end": 11.48, "probability": 0.99755859375 }, { "word": "両", "start": 11.48, "end": 11.86, "probability": 0.9990234375 }, { "word": "側", "start": 11.86, "end": 12.0, "probability": 1.0 }, { "word": "の", "start": 12.0, "end": 12.12, "probability": 1.0 }, { "word": "ゴ", "start": 12.12, "end": 12.3, "probability": 0.98828125 }, { "word": "ム", "start": 12.3, "end": 12.34, "probability": 1.0 }, { "word": "缶", "start": 12.34, "end": 12.5, "probability": 0.9814453125 }, { "word": "を", "start": 12.5, "end": 12.62, "probability": 1.0 }, { "word": "吹", "start": 12.62, "end": 12.74, "probability": 0.9111328125 }, { "word": "いて", "start": 12.74, "end": 12.84, "probability": 1.0 }, { "word": "ください", "start": 12.84, "end": 13.44, "probability": 0.99853515625 } ] }, { "id": 2, "seek": 0, "start": 13.44, "end": 16.28, "text": "膨らみが足りない場合は両側のゴム缶をご覧ください", "tokens": [ 51030, 39708, 101, 5154, 11362, 5142, 37236, 5095, 9311, 21446, 14245, 3065, 940, 94, 7437, 112, 2972, 39780, 32026, 38109, 114, 5998, 9991, 2888, 100, 25079, 51231 ], "temperature": 0.0, "avg_logprob": NaN, "compression_ratio": 1.68125, "no_speech_prob": 0.14235487580299377, "words": [ { "word": "膨", "start": 13.44, "end": 13.74, "probability": 0.505859375 }, { "word": "ら", "start": 13.74, "end": 13.86, "probability": 0.99755859375 }, { "word": "み", "start": 13.86, "end": 14.12, "probability": 0.9990234375 }, { "word": "が", "start": 14.12, "end": 14.22, "probability": 0.99951171875 }, { "word": "足", "start": 14.22, "end": 14.34, "probability": 1.0 }, { "word": "り", "start": 14.34, "end": 14.34, "probability": 1.0 }, { "word": "ない", "start": 14.34, "end": 14.44, "probability": 0.99951171875 }, { "word": "場", "start": 14.44, "end": 14.48, "probability": 1.0 }, { "word": "合", "start": 14.48, "end": 14.68, "probability": 1.0 }, { "word": "は", "start": 14.68, "end": 14.92, "probability": 0.998046875 }, { "word": "両", "start": 14.92, "end": 15.14, "probability": 0.9921875 }, { "word": "側", "start": 15.14, "end": 15.26, "probability": 1.0 }, { "word": "の", "start": 15.26, "end": 15.26, "probability": 0.99951171875 }, { "word": "ゴ", "start": 15.26, "end": 15.26, "probability": 0.99560546875 }, { "word": "ム", "start": 15.26, "end": 15.26, "probability": 1.0 }, { "word": "缶", "start": 15.26, "end": 15.26, "probability": 0.9921875 }, { "word": "を", "start": 15.26, "end": 15.68, "probability": 1.0 }, { "word": "ご", "start": 15.68, "end": 15.68, "probability": 0.050445556640625 }, { "word": "覧", "start": 15.68, "end": 15.82, "probability": 0.994140625 }, { "word": "ください", "start": 15.82, "end": 16.28, "probability": 0.99951171875 } ] }, { "id": 3, "seek": 0, "start": 16.28, "end": 19.400000000000002, "text": "ぜひご覧ください", "tokens": [ 51231, 20258, 26950, 9991, 2888, 100, 25079, 51335 ], "temperature": 0.0, "avg_logprob": NaN, "compression_ratio": 1.68125, "no_speech_prob": 0.14235487580299377, "words": [ { "word": "ぜ", "start": 16.28, "end": 16.88, "probability": 0.0302886962890625 }, { "word": "ひ", "start": 16.88, "end": 17.06, "probability": 0.73583984375 }, { "word": "ご", "start": 17.06, "end": 17.26, "probability": 3.93986701965332e-05 }, { "word": "覧", "start": 17.26, "end": 17.26, "probability": 0.9873046875 }, { "word": "ください!!!!!", "start": 17.26, "end": 19.400000000000002, "probability": 0.99609375 } ] }, { "id": 4, "seek": 0, "start": 19.400000000000002, "end": 7.38, "text": "!", "tokens": [ 51335, 0, 50734 ], "temperature": 0.0, "avg_logprob": NaN, "compression_ratio": 1.68125, "no_speech_prob": 0.14235487580299377, "words": [] }, { "id": 5, "seek": 0, "start": 7.38, "end": 7.38, "text": "", "tokens": [], "temperature": 0.0, "avg_logprob": NaN, "compression_ratio": 1.68125, "no_speech_prob": 0.14235487580299377, "words": [] }, { "id": 6, "seek": 0, "start": 7.38, "end": 9.34, "text": "!", "tokens": [ 50734, 0, 50832 ], "temperature": 0.0, "avg_logprob": NaN, "compression_ratio": 1.68125, "no_speech_prob": 0.14235487580299377, "words": [] }, { "id": 7, "seek": 0, "start": 9.34, "end": 9.34, "text": "", "tokens": [], "temperature": 0.0, "avg_logprob": NaN, "compression_ratio": 1.68125, "no_speech_prob": 0.14235487580299377, "words": [] }, { "id": 8, "seek": 0, "start": 9.34, "end": 23.34, "text": "!", "tokens": [ 50832, 0, 51532 ], "temperature": 0.0, "avg_logprob": NaN, "compression_ratio": 1.68125, "no_speech_prob": 0.14235487580299377, "words": [] } ], "language": "ja", "duration": 24.6 }, "status": "COMPLETE", "rejected_alternative": { "text": "非常口または翼の上で前の引き手を強く引いて膨らませてください。 Inflate the life vest at the door or on the wing by pulling both knobs. 膨らみが足りない場合は両側のゴム管を吹いてください。 If you need more air, blow into the tubes. 坐席ポケットの安全のしおりをご覧ください。 Please look at the safety instructions card in your seat pocket.", "chunks": [ { "text": "非常口または翼の上で前の引き手を強く引いて膨らませてください。 Inflate the life vest at the door or on the wing by pulling both knobs. 膨らみが足りない場合は両側のゴム管を吹いてください。 If you need more air, blow into the tubes. 坐席ポケットの安全のしおりをご覧ください。 Please look at the safety instructions card in your seat pocket.", "start": 0.0, "end": 24.6, "chunk_index": 0, "language": "Japanese", "finish_reason": "eos", "truncated": false, "generated_tokens": 97, "max_new_tokens": 328 } ], "detected_language": "Japanese", "segments": [ { "start": 379.4, "end": 404.0, "text": "非常口または翼の上で前の引き手を強く引いて膨らませてください。 Inflate the life vest at the door or on the wing by pulling both knobs. 膨らみが足りない場合は両側のゴム管を吹いてください。 If you need more air, blow into the tubes. 坐席ポケットの安全のしおりをご覧ください。 Please look at the safety instructions card in your seat pocket." } ], "start": 379.4, "end": 404.0, "engine": "mlx-qwen3-asr", "language": "ja" } }, "raw_word_timestamps": [ { "word": "膨", "start": 389.34, "end": 389.65999999999997, "probability": 0.8486328125 }, { "word": "ら", "start": 389.65999999999997, "end": 389.71999999999997, "probability": 0.998046875 }, { "word": "み", "start": 389.71999999999997, "end": 389.82, "probability": 1.0 }, { "word": "が", "start": 389.82, "end": 389.94, "probability": 0.9990234375 }, { "word": "足", "start": 389.94, "end": 390.03999999999996, "probability": 0.99951171875 }, { "word": "り", "start": 390.03999999999996, "end": 390.15999999999997, "probability": 1.0 }, { "word": "ない", "start": 390.15999999999997, "end": 390.29999999999995, "probability": 1.0 }, { "word": "場", "start": 390.29999999999995, "end": 390.47999999999996, "probability": 1.0 }, { "word": "合", "start": 390.47999999999996, "end": 390.62, "probability": 1.0 }, { "word": "は", "start": 390.62, "end": 390.79999999999995, "probability": 0.99951171875 }, { "word": "両", "start": 390.79999999999995, "end": 391.2, "probability": 0.9912109375 }, { "word": "側", "start": 391.2, "end": 391.38, "probability": 1.0 }, { "word": "の", "start": 391.38, "end": 391.5, "probability": 1.0 }, { "word": "ゴ", "start": 391.5, "end": 391.59999999999997, "probability": 0.99853515625 }, { "word": "ム", "start": 391.59999999999997, "end": 391.71999999999997, "probability": 0.99951171875 }, { "word": "缶", "start": 391.71999999999997, "end": 391.85999999999996, "probability": 0.9208984375 }, { "word": "を", "start": 391.85999999999996, "end": 392.0, "probability": 1.0 }, { "word": "吹", "start": 392.0, "end": 392.12, "probability": 0.8564453125 }, { "word": "いて", "start": 392.12, "end": 392.23999999999995, "probability": 1.0 }, { "word": "ください", "start": 392.23999999999995, "end": 392.67999999999995, "probability": 0.99951171875 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "膨らみが足りない場合は両側のゴム管を吹いてください", "review_status": "STRONG_RESTORATION", "classification": "STRONG_RESTORATION", "confidence": 0.9, "correction_reason": "ASR 'ゴム缶' (can) is a phonetic error for 'ゴム管' (tube), standard life vest instruction." }, { "start": 395.96, "end": 398.78, "text": "座席ポケットの安全のしおりをご覧ください", "raw_asr_text": "座席ポケットの安全の修理をご覧ください", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 20, "words": [], "asr_metrics": { "avg_logprob": -0.11740236918131511, "compression_ratio": 1.3703703703703705, "no_speech_prob": 3.623901526084161e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "非常口または翼の上で前の引き手を強く引いて膨らませてください膨らみが足りない場合は両側のゴム缶を吹いてください座席ポケットの安全の修理をご覧ください" }, { "language": "en", "text": "On the right side of the seat, the front side of the seat is a very strong one. If you need more air, blow into the tubes. Please look at the safety instructions card in your seat pocket." } ], "id": 65, "review_annotation": "", "row_id": 65, "bounded_row_id": 65, "parent_row_ids": [ 65 ], "parent_row_id": 65, "packet_id": 20, "LID_evidence": [ { "region_id": 50, "probabilities": { "ja": 0.2625669730619443, "zh": 0.004405638382498485, "en": 0.5449164732255869 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 50 ], "language_confidence": 0.2625669730619443, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 379.4, "packet_end": 404.0, "raw_text": "座席ポケットの安全の修理をご覧ください", "selected_text": "座席ポケットの安全のしおりをご覧ください", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 65 ], "packet_id": 20, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "座席ポケットの安全の修理をご覧ください", "after": "座席ポケットの安全のしおりをご覧ください", "classification": "STRONG_RESTORATION", "reason": "ASR '修理' (repair) is a phonetic error for 'しおり' (leaflet/booklet), standard safety instruction." } ], "language": "ja", "packet_ids": [ 20 ], "index": 65, "independent_ASR": { "engine": "kotoba-whisper", "model_path": "/Users/sagawa/AI_WORK/cache/huggingface/hub/models--kaiinui--kotoba-whisper-v2.0-mlx/snapshots/1a5ac6e52394898ef8996212e849d964e0dfb9ae", "language": "ja", "audio_interval": [ 379.4, 404.0 ], "target_interval": [ 380.2, 403.2 ], "text": "非常口または翼の上で前の引き手を強く引いて膨らませてくださいフクラミが足りない場合は両側のゴム缶を吹いてください膨らみが足りない場合は両側のゴム缶をご覧くださいぜひご覧ください!!!!!", "words": [ { "word": "非常", "start": 380.2, "end": 380.56, "probability": 0.59423828125 }, { "word": "口", "start": 380.56, "end": 380.88, "probability": 0.9736328125 }, { "word": "また", "start": 380.88, "end": 381.17999999999995, "probability": 0.99609375 }, { "word": "は", "start": 381.17999999999995, "end": 381.34, "probability": 0.99755859375 }, { "word": "翼", "start": 381.34, "end": 381.5, "probability": 0.99169921875 }, { "word": "の", "start": 381.5, "end": 381.71999999999997, "probability": 0.99951171875 }, { "word": "上", "start": 381.71999999999997, "end": 381.9, "probability": 0.99951171875 }, { "word": "で", "start": 381.9, "end": 382.15999999999997, "probability": 0.99951171875 }, { "word": "前", "start": 382.15999999999997, "end": 382.52, "probability": 0.951171875 }, { "word": "の", "start": 382.52, "end": 382.7, "probability": 0.9990234375 }, { "word": "引", "start": 382.7, "end": 382.84, "probability": 0.98876953125 }, { "word": "き", "start": 382.84, "end": 382.84, "probability": 0.99951171875 }, { "word": "手", "start": 382.84, "end": 382.94, "probability": 0.99951171875 }, { "word": "を", "start": 382.94, "end": 383.03999999999996, "probability": 0.9990234375 }, { "word": "強", "start": 383.03999999999996, "end": 383.23999999999995, "probability": 0.99072265625 }, { "word": "く", "start": 383.23999999999995, "end": 383.4, "probability": 1.0 }, { "word": "引", "start": 383.4, "end": 383.56, "probability": 0.76904296875 }, { "word": "いて", "start": 383.56, "end": 383.7, "probability": 0.99951171875 }, { "word": "膨", "start": 383.7, "end": 383.97999999999996, "probability": 0.99072265625 }, { "word": "ら", "start": 383.97999999999996, "end": 384.03999999999996, "probability": 1.0 }, { "word": "ませ", "start": 384.03999999999996, "end": 384.26, "probability": 0.99951171875 }, { "word": "て", "start": 384.26, "end": 384.35999999999996, "probability": 0.99951171875 }, { "word": "ください", "start": 384.35999999999996, "end": 384.85999999999996, "probability": 0.998046875 }, { "word": "フ", "start": 384.85999999999996, "end": 385.14, "probability": 0.067138671875 }, { "word": "ク", "start": 385.14, "end": 385.21999999999997, "probability": 0.74072265625 }, { "word": "ラ", "start": 385.21999999999997, "end": 385.34, "probability": 0.9970703125 }, { "word": "ミ", "start": 385.34, "end": 385.52, "probability": 0.94482421875 }, { "word": "が", "start": 385.52, "end": 385.52, "probability": 0.99951171875 }, { "word": "足", "start": 385.52, "end": 385.91999999999996, "probability": 1.0 }, { "word": "り", "start": 385.91999999999996, "end": 386.09999999999997, "probability": 1.0 }, { "word": "ない", "start": 386.09999999999997, "end": 386.34, "probability": 0.99951171875 }, { "word": "場", "start": 386.34, "end": 386.62, "probability": 0.99951171875 }, { "word": "合", "start": 386.62, "end": 390.85999999999996, "probability": 1.0 }, { "word": "は", "start": 390.85999999999996, "end": 390.88, "probability": 0.99755859375 }, { "word": "両", "start": 390.88, "end": 391.26, "probability": 0.9990234375 }, { "word": "側", "start": 391.26, "end": 391.4, "probability": 1.0 }, { "word": "の", "start": 391.4, "end": 391.52, "probability": 1.0 }, { "word": "ゴ", "start": 391.52, "end": 391.7, "probability": 0.98828125 }, { "word": "ム", "start": 391.7, "end": 391.73999999999995, "probability": 1.0 }, { "word": "缶", "start": 391.73999999999995, "end": 391.9, "probability": 0.9814453125 }, { "word": "を", "start": 391.9, "end": 392.02, "probability": 1.0 }, { "word": "吹", "start": 392.02, "end": 392.14, "probability": 0.9111328125 }, { "word": "いて", "start": 392.14, "end": 392.23999999999995, "probability": 1.0 }, { "word": "ください", "start": 392.23999999999995, "end": 392.84, "probability": 0.99853515625 }, { "word": "膨", "start": 392.84, "end": 393.14, "probability": 0.505859375 }, { "word": "ら", "start": 393.14, "end": 393.26, "probability": 0.99755859375 }, { "word": "み", "start": 393.26, "end": 393.52, "probability": 0.9990234375 }, { "word": "が", "start": 393.52, "end": 393.62, "probability": 0.99951171875 }, { "word": "足", "start": 393.62, "end": 393.73999999999995, "probability": 1.0 }, { "word": "り", "start": 393.73999999999995, "end": 393.73999999999995, "probability": 1.0 }, { "word": "ない", "start": 393.73999999999995, "end": 393.84, "probability": 0.99951171875 }, { "word": "場", "start": 393.84, "end": 393.88, "probability": 1.0 }, { "word": "合", "start": 393.88, "end": 394.08, "probability": 1.0 }, { "word": "は", "start": 394.08, "end": 394.32, "probability": 0.998046875 }, { "word": "両", "start": 394.32, "end": 394.53999999999996, "probability": 0.9921875 }, { "word": "側", "start": 394.53999999999996, "end": 394.65999999999997, "probability": 1.0 }, { "word": "の", "start": 394.65999999999997, "end": 394.65999999999997, "probability": 0.99951171875 }, { "word": "ゴ", "start": 394.65999999999997, "end": 394.65999999999997, "probability": 0.99560546875 }, { "word": "ム", "start": 394.65999999999997, "end": 394.65999999999997, "probability": 1.0 }, { "word": "缶", "start": 394.65999999999997, "end": 394.65999999999997, "probability": 0.9921875 }, { "word": "を", "start": 394.65999999999997, "end": 395.08, "probability": 1.0 }, { "word": "ご", "start": 395.08, "end": 395.08, "probability": 0.050445556640625 }, { "word": "覧", "start": 395.08, "end": 395.21999999999997, "probability": 0.994140625 }, { "word": "ください", "start": 395.21999999999997, "end": 395.67999999999995, "probability": 0.99951171875 }, { "word": "ぜ", "start": 395.67999999999995, "end": 396.28, "probability": 0.0302886962890625 }, { "word": "ひ", "start": 396.28, "end": 396.46, "probability": 0.73583984375 }, { "word": "ご", "start": 396.46, "end": 396.65999999999997, "probability": 3.93986701965332e-05 }, { "word": "覧", "start": 396.65999999999997, "end": 396.65999999999997, "probability": 0.9873046875 }, { "word": "ください!!!!!", "start": 396.65999999999997, "end": 398.79999999999995, "probability": 0.99609375 } ], "raw_result": { "text": "非常口または翼の上で前の引き手を強く引いて膨らませてくださいフクラミが足りない場合は両側のゴム缶を吹いてください膨らみが足りない場合は両側のゴム缶をご覧くださいぜひご覧ください!!!", "segments": [ { "id": 0, "seek": 0, "start": 0.7999999999999996, "end": 5.46, "text": "非常口または翼の上で前の引き手を強く引いて膨らませてください", "tokens": [ 50365, 14392, 18144, 24712, 3065, 18573, 120, 2972, 5708, 2474, 8945, 2972, 41301, 7016, 11389, 5998, 22752, 6134, 41301, 18549, 39708, 101, 5154, 28742, 2996, 25079, 50633 ], "temperature": 0.0, "avg_logprob": NaN, "compression_ratio": 1.68125, "no_speech_prob": 0.14235487580299377, "words": [ { "word": "非常", "start": 0.7999999999999996, "end": 1.16, "probability": 0.59423828125 }, { "word": "口", "start": 1.16, "end": 1.48, "probability": 0.9736328125 }, { "word": "また", "start": 1.48, "end": 1.78, "probability": 0.99609375 }, { "word": "は", "start": 1.78, "end": 1.94, "probability": 0.99755859375 }, { "word": "翼", "start": 1.94, "end": 2.1, "probability": 0.99169921875 }, { "word": "の", "start": 2.1, "end": 2.32, "probability": 0.99951171875 }, { "word": "上", "start": 2.32, "end": 2.5, "probability": 0.99951171875 }, { "word": "で", "start": 2.5, "end": 2.76, "probability": 0.99951171875 }, { "word": "前", "start": 2.76, "end": 3.12, "probability": 0.951171875 }, { "word": "の", "start": 3.12, "end": 3.3, "probability": 0.9990234375 }, { "word": "引", "start": 3.3, "end": 3.44, "probability": 0.98876953125 }, { "word": "き", "start": 3.44, "end": 3.44, "probability": 0.99951171875 }, { "word": "手", "start": 3.44, "end": 3.54, "probability": 0.99951171875 }, { "word": "を", "start": 3.54, "end": 3.64, "probability": 0.9990234375 }, { "word": "強", "start": 3.64, "end": 3.84, "probability": 0.99072265625 }, { "word": "く", "start": 3.84, "end": 4.0, "probability": 1.0 }, { "word": "引", "start": 4.0, "end": 4.16, "probability": 0.76904296875 }, { "word": "いて", "start": 4.16, "end": 4.3, "probability": 0.99951171875 }, { "word": "膨", "start": 4.3, "end": 4.58, "probability": 0.99072265625 }, { "word": "ら", "start": 4.58, "end": 4.64, "probability": 1.0 }, { "word": "ませ", "start": 4.64, "end": 4.86, "probability": 0.99951171875 }, { "word": "て", "start": 4.86, "end": 4.96, "probability": 0.99951171875 }, { "word": "ください", "start": 4.96, "end": 5.46, "probability": 0.998046875 } ] }, { "id": 1, "seek": 0, "start": 5.46, "end": 13.44, "text": "フクラミが足りない場合は両側のゴム缶を吹いてください", "tokens": [ 50633, 17320, 10825, 11353, 23196, 5142, 37236, 5095, 9311, 21446, 14245, 3065, 940, 94, 7437, 112, 2972, 39780, 32026, 38109, 114, 5998, 2392, 117, 18549, 25079, 51030 ], "temperature": 0.0, "avg_logprob": NaN, "compression_ratio": 1.68125, "no_speech_prob": 0.14235487580299377, "words": [ { "word": "フ", "start": 5.46, "end": 5.74, "probability": 0.067138671875 }, { "word": "ク", "start": 5.74, "end": 5.82, "probability": 0.74072265625 }, { "word": "ラ", "start": 5.82, "end": 5.94, "probability": 0.9970703125 }, { "word": "ミ", "start": 5.94, "end": 6.12, "probability": 0.94482421875 }, { "word": "が", "start": 6.12, "end": 6.12, "probability": 0.99951171875 }, { "word": "足", "start": 6.12, "end": 6.52, "probability": 1.0 }, { "word": "り", "start": 6.52, "end": 6.7, "probability": 1.0 }, { "word": "ない", "start": 6.7, "end": 6.94, "probability": 0.99951171875 }, { "word": "場", "start": 6.94, "end": 7.22, "probability": 0.99951171875 }, { "word": "合", "start": 7.22, "end": 11.46, "probability": 1.0 }, { "word": "は", "start": 11.46, "end": 11.48, "probability": 0.99755859375 }, { "word": "両", "start": 11.48, "end": 11.86, "probability": 0.9990234375 }, { "word": "側", "start": 11.86, "end": 12.0, "probability": 1.0 }, { "word": "の", "start": 12.0, "end": 12.12, "probability": 1.0 }, { "word": "ゴ", "start": 12.12, "end": 12.3, "probability": 0.98828125 }, { "word": "ム", "start": 12.3, "end": 12.34, "probability": 1.0 }, { "word": "缶", "start": 12.34, "end": 12.5, "probability": 0.9814453125 }, { "word": "を", "start": 12.5, "end": 12.62, "probability": 1.0 }, { "word": "吹", "start": 12.62, "end": 12.74, "probability": 0.9111328125 }, { "word": "いて", "start": 12.74, "end": 12.84, "probability": 1.0 }, { "word": "ください", "start": 12.84, "end": 13.44, "probability": 0.99853515625 } ] }, { "id": 2, "seek": 0, "start": 13.44, "end": 16.28, "text": "膨らみが足りない場合は両側のゴム缶をご覧ください", "tokens": [ 51030, 39708, 101, 5154, 11362, 5142, 37236, 5095, 9311, 21446, 14245, 3065, 940, 94, 7437, 112, 2972, 39780, 32026, 38109, 114, 5998, 9991, 2888, 100, 25079, 51231 ], "temperature": 0.0, "avg_logprob": NaN, "compression_ratio": 1.68125, "no_speech_prob": 0.14235487580299377, "words": [ { "word": "膨", "start": 13.44, "end": 13.74, "probability": 0.505859375 }, { "word": "ら", "start": 13.74, "end": 13.86, "probability": 0.99755859375 }, { "word": "み", "start": 13.86, "end": 14.12, "probability": 0.9990234375 }, { "word": "が", "start": 14.12, "end": 14.22, "probability": 0.99951171875 }, { "word": "足", "start": 14.22, "end": 14.34, "probability": 1.0 }, { "word": "り", "start": 14.34, "end": 14.34, "probability": 1.0 }, { "word": "ない", "start": 14.34, "end": 14.44, "probability": 0.99951171875 }, { "word": "場", "start": 14.44, "end": 14.48, "probability": 1.0 }, { "word": "合", "start": 14.48, "end": 14.68, "probability": 1.0 }, { "word": "は", "start": 14.68, "end": 14.92, "probability": 0.998046875 }, { "word": "両", "start": 14.92, "end": 15.14, "probability": 0.9921875 }, { "word": "側", "start": 15.14, "end": 15.26, "probability": 1.0 }, { "word": "の", "start": 15.26, "end": 15.26, "probability": 0.99951171875 }, { "word": "ゴ", "start": 15.26, "end": 15.26, "probability": 0.99560546875 }, { "word": "ム", "start": 15.26, "end": 15.26, "probability": 1.0 }, { "word": "缶", "start": 15.26, "end": 15.26, "probability": 0.9921875 }, { "word": "を", "start": 15.26, "end": 15.68, "probability": 1.0 }, { "word": "ご", "start": 15.68, "end": 15.68, "probability": 0.050445556640625 }, { "word": "覧", "start": 15.68, "end": 15.82, "probability": 0.994140625 }, { "word": "ください", "start": 15.82, "end": 16.28, "probability": 0.99951171875 } ] }, { "id": 3, "seek": 0, "start": 16.28, "end": 19.400000000000002, "text": "ぜひご覧ください", "tokens": [ 51231, 20258, 26950, 9991, 2888, 100, 25079, 51335 ], "temperature": 0.0, "avg_logprob": NaN, "compression_ratio": 1.68125, "no_speech_prob": 0.14235487580299377, "words": [ { "word": "ぜ", "start": 16.28, "end": 16.88, "probability": 0.0302886962890625 }, { "word": "ひ", "start": 16.88, "end": 17.06, "probability": 0.73583984375 }, { "word": "ご", "start": 17.06, "end": 17.26, "probability": 3.93986701965332e-05 }, { "word": "覧", "start": 17.26, "end": 17.26, "probability": 0.9873046875 }, { "word": "ください!!!!!", "start": 17.26, "end": 19.400000000000002, "probability": 0.99609375 } ] }, { "id": 4, "seek": 0, "start": 19.400000000000002, "end": 7.38, "text": "!", "tokens": [ 51335, 0, 50734 ], "temperature": 0.0, "avg_logprob": NaN, "compression_ratio": 1.68125, "no_speech_prob": 0.14235487580299377, "words": [] }, { "id": 5, "seek": 0, "start": 7.38, "end": 7.38, "text": "", "tokens": [], "temperature": 0.0, "avg_logprob": NaN, "compression_ratio": 1.68125, "no_speech_prob": 0.14235487580299377, "words": [] }, { "id": 6, "seek": 0, "start": 7.38, "end": 9.34, "text": "!", "tokens": [ 50734, 0, 50832 ], "temperature": 0.0, "avg_logprob": NaN, "compression_ratio": 1.68125, "no_speech_prob": 0.14235487580299377, "words": [] }, { "id": 7, "seek": 0, "start": 9.34, "end": 9.34, "text": "", "tokens": [], "temperature": 0.0, "avg_logprob": NaN, "compression_ratio": 1.68125, "no_speech_prob": 0.14235487580299377, "words": [] }, { "id": 8, "seek": 0, "start": 9.34, "end": 23.34, "text": "!", "tokens": [ 50832, 0, 51532 ], "temperature": 0.0, "avg_logprob": NaN, "compression_ratio": 1.68125, "no_speech_prob": 0.14235487580299377, "words": [] } ], "language": "ja", "duration": 24.6 }, "status": "COMPLETE", "rejected_alternative": { "text": "非常口または翼の上で前の引き手を強く引いて膨らませてください。 Inflate the life vest at the door or on the wing by pulling both knobs. 膨らみが足りない場合は両側のゴム管を吹いてください。 If you need more air, blow into the tubes. 坐席ポケットの安全のしおりをご覧ください。 Please look at the safety instructions card in your seat pocket.", "chunks": [ { "text": "非常口または翼の上で前の引き手を強く引いて膨らませてください。 Inflate the life vest at the door or on the wing by pulling both knobs. 膨らみが足りない場合は両側のゴム管を吹いてください。 If you need more air, blow into the tubes. 坐席ポケットの安全のしおりをご覧ください。 Please look at the safety instructions card in your seat pocket.", "start": 0.0, "end": 24.6, "chunk_index": 0, "language": "Japanese", "finish_reason": "eos", "truncated": false, "generated_tokens": 97, "max_new_tokens": 328 } ], "detected_language": "Japanese", "segments": [ { "start": 379.4, "end": 404.0, "text": "非常口または翼の上で前の引き手を強く引いて膨らませてください。 Inflate the life vest at the door or on the wing by pulling both knobs. 膨らみが足りない場合は両側のゴム管を吹いてください。 If you need more air, blow into the tubes. 坐席ポケットの安全のしおりをご覧ください。 Please look at the safety instructions card in your seat pocket." } ], "start": 379.4, "end": 404.0, "engine": "mlx-qwen3-asr", "language": "ja" } }, "raw_word_timestamps": [ { "word": "座", "start": 395.96, "end": 396.28, "probability": 0.966796875 }, { "word": "席", "start": 396.28, "end": 396.46, "probability": 0.99951171875 }, { "word": "ポ", "start": 396.46, "end": 396.62, "probability": 0.9990234375 }, { "word": "ケ", "start": 396.62, "end": 396.71999999999997, "probability": 1.0 }, { "word": "ット", "start": 396.71999999999997, "end": 396.88, "probability": 1.0 }, { "word": "の", "start": 396.88, "end": 397.06, "probability": 0.99951171875 }, { "word": "安全", "start": 397.06, "end": 397.46, "probability": 0.99951171875 }, { "word": "の", "start": 397.46, "end": 397.65999999999997, "probability": 0.98974609375 }, { "word": "修", "start": 397.65999999999997, "end": 397.82, "probability": 0.9130859375 }, { "word": "理", "start": 397.82, "end": 397.94, "probability": 0.998046875 }, { "word": "を", "start": 397.94, "end": 398.08, "probability": 1.0 }, { "word": "ご", "start": 398.08, "end": 398.26, "probability": 0.99951171875 }, { "word": "覧", "start": 398.26, "end": 398.35999999999996, "probability": 0.99951171875 }, { "word": "ください", "start": 398.35999999999996, "end": 398.78, "probability": 0.998046875 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "座席ポケットの安全のしおりをご覧ください", "review_status": "STRONG_RESTORATION", "classification": "STRONG_RESTORATION", "confidence": 0.9, "correction_reason": "ASR '修理' (repair) is a phonetic error for 'しおり' (leaflet/booklet), standard safety instruction." }, { "start": 403.9, "end": 410.15999999999997, "text": "If you have any questions, please contact the cabin attendant.", "raw_asr_text": "If you have any questions, please contact the cabin attendant.", "source_language": "en", "source_lang": "en", "asr_packet_id": 21, "words": [], "asr_metrics": { "avg_logprob": -0.4728501955668131, "compression_ratio": 0.9393939393939394, "no_speech_prob": 1.128272544836939e-10 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "zh", "text": "如果有質疑有問題的話,請問客室上務員如果有問題的話,請問客室上" }, { "language": "en", "text": "If you have any questions, please contact the cabin attendant." }, { "language": "ja", "text": "ご質問がございましたら客室乗務員にお尋ねくださいもし問題があるなら、カバンの探索を通してください" } ], "id": 66, "review_annotation": "", "row_id": 66, "bounded_row_id": 66, "parent_row_ids": [ 66 ], "parent_row_id": 66, "packet_id": 21, "LID_evidence": [ { "region_id": 51, "probabilities": { "ja": 0.13786083865644677, "zh": 0.16741397898710733, "en": 0.3273087504785508 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 51 ], "language_confidence": 0.3273087504785508, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 403.09999999999997, "packet_end": 411.6, "raw_text": "If you have any questions, please contact the cabin attendant.", "selected_text": "If you have any questions, please contact the cabin attendant.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 66 ], "packet_id": 21, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "If you have any questions, please contact the cabin attendant.", "after": "If you have any questions, please contact the cabin attendant.", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard English safety announcement; matches ASR and context perfectly." } ], "language": "en", "packet_ids": [ 21 ], "index": 66, "source_text": "If you have any questions, please contact the cabin attendant.", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Standard English safety announcement; matches ASR and context perfectly.", "raw_word_timestamps": [ { "word": " If", "start": 403.09999999999997, "end": 404.09999999999997, "probability": 0.7841796875 }, { "word": " you", "start": 404.09999999999997, "end": 404.2, "probability": 0.8671875 }, { "word": " have", "start": 404.2, "end": 408.11999999999995, "probability": 0.9453125 }, { "word": " any", "start": 408.11999999999995, "end": 408.28, "probability": 0.951171875 }, { "word": " questions,", "start": 408.28, "end": 408.64, "probability": 0.9697265625 }, { "word": " please", "start": 409.08, "end": 409.34, "probability": 0.9931640625 }, { "word": " contact", "start": 409.34, "end": 409.64, "probability": 0.98681640625 }, { "word": " the", "start": 409.64, "end": 409.78, "probability": 0.8134765625 }, { "word": " cabin", "start": 409.78, "end": 409.93999999999994, "probability": 0.384033203125 }, { "word": " attendant.", "start": 409.93999999999994, "end": 410.15999999999997, "probability": 0.8994140625 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 418.64, "end": 422.98, "text": "皆様、入陸いたします。シートベルトをもう一度お確かめください。", "raw_asr_text": "皆様、入陸いたします。シートベルトをもう一度お確かめください。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 22, "words": [ { "word": "皆", "start": 418.64, "end": 418.96000000000004, "probability": 0.9326171875 }, { "word": "様、", "start": 418.96000000000004, "end": 419.28000000000003, "probability": 0.63330078125 }, { "word": "入", "start": 419.44, "end": 419.52000000000004, "probability": 0.427734375 }, { "word": "陸", "start": 419.52000000000004, "end": 419.78000000000003, "probability": 0.99462890625 }, { "word": "いた", "start": 419.78000000000003, "end": 419.92, "probability": 0.9873046875 }, { "word": "します。", "start": 419.92, "end": 420.28000000000003, "probability": 1.0 }, { "word": "シ", "start": 420.58, "end": 420.86, "probability": 0.962890625 }, { "word": "ート", "start": 420.86, "end": 421.02000000000004, "probability": 1.0 }, { "word": "ベ", "start": 421.02000000000004, "end": 421.18, "probability": 0.9892578125 }, { "word": "ル", "start": 421.18, "end": 421.24, "probability": 0.9990234375 }, { "word": "ト", "start": 421.24, "end": 421.38, "probability": 0.99951171875 }, { "word": "を", "start": 421.38, "end": 421.54, "probability": 1.0 }, { "word": "もう", "start": 421.54, "end": 421.7, "probability": 0.99951171875 }, { "word": "一", "start": 421.7, "end": 421.88, "probability": 1.0 }, { "word": "度", "start": 421.88, "end": 422.1, "probability": 1.0 }, { "word": "お", "start": 422.1, "end": 422.26, "probability": 0.86962890625 }, { "word": "確", "start": 422.26, "end": 422.36, "probability": 0.998046875 }, { "word": "か", "start": 422.36, "end": 422.54, "probability": 0.99951171875 }, { "word": "め", "start": 422.54, "end": 422.64, "probability": 0.9990234375 }, { "word": "ください。", "start": 422.64, "end": 422.98, "probability": 0.99755859375 } ], "asr_metrics": { "avg_logprob": -0.15221533951935945, "compression_ratio": 1.0108695652173914, "no_speech_prob": 5.058180437000903e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "皆様、入陸いたします。シートベルトをもう一度お確かめください。" } ], "id": 67, "review_annotation": "", "row_id": 67, "bounded_row_id": 67, "parent_row_ids": [ 67 ], "parent_row_id": 67, "packet_id": 22, "LID_evidence": [ { "region_id": 52, "probabilities": { "ja": 0.9665452539920807, "zh": 0.0018321676718187519, "en": 0.017964317608857527 }, "status": "STABLE" } ], "VAD_region_ids": [ 52 ], "language_confidence": 0.9665452539920807, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 417.8, "packet_end": 424.40000000000003, "raw_text": "皆様、入陸いたします。シートベルトをもう一度お確かめください。", "selected_text": "皆様、入陸いたします。シートベルトをもう一度お確かめください。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 67 ], "packet_id": 22, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "皆様、入陸いたします。シートベルトをもう一度お確かめください。", "after": "皆様、入陸いたします。シートベルトをもう一度お確かめください。", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Transcription matches standard announcement phrasing and ASR output." } ], "language": "ja", "packet_ids": [ 22 ], "index": 67, "independent_ASR": { "text": "皆様、入力いたします。シートベルトをもう一度お確かめください。", "chunks": [ { "text": "皆様、入力いたします。シートベルトをもう一度お確かめください。", "start": 0.0, "end": 6.6, "chunk_index": 0, "language": "Japanese", "finish_reason": "eos", "truncated": false, "generated_tokens": 19, "max_new_tokens": 128 } ], "detected_language": "Japanese", "segments": [ { "start": 417.8, "end": 424.40000000000003, "text": "皆様、入力いたします。シートベルトをもう一度お確かめください。" } ], "start": 417.8, "end": 424.40000000000003, "engine": "mlx-qwen3-asr", "language": "ja" }, "source_text": "皆様、入陸いたします。シートベルトをもう一度お確かめください。", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Transcription matches standard announcement phrasing and ASR output." }, { "start": 424.2, "end": 431.41999999999996, "text": "Ladies and gentlemen, we are now ready for take off. Please make sure that your seat belt is security fastened. Thank you.", "raw_asr_text": "Ladies and gentlemen, we are now ready for take off. Please make sure that your seat belt is security fastened. Thank you.", "source_language": "en", "source_lang": "en", "asr_packet_id": 23, "words": [], "asr_metrics": { "avg_logprob": -0.32106723785400393, "compression_ratio": 1.1844660194174756, "no_speech_prob": 7.521699235679691e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "en", "text": "Ladies and gentlemen, we are now ready for take off. Please make sure that your seat belt is security fastened. Thank you." } ], "id": 68, "review_annotation": "", "row_id": 68, "bounded_row_id": 68, "parent_row_ids": [ 68 ], "parent_row_id": 68, "packet_id": 23, "LID_evidence": [ { "region_id": 53, "probabilities": { "ja": 0.2589338409248739, "zh": 0.015075960149988531, "en": 0.5788973767310381 }, "status": "STABLE" }, { "region_id": 54, "probabilities": { "ja": 0.00985506779074462, "zh": 0.010529290347929217, "en": 0.8760179645485349 }, "status": "STABLE" } ], "VAD_region_ids": [ 53, 54 ], "language_confidence": 0.8760179645485349, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 423.4, "packet_end": 432.6, "raw_text": "Ladies and gentlemen, we are now ready for take off. Please make sure that your seat belt is security fastened. Thank you.", "selected_text": "Ladies and gentlemen, we are now ready for take off. Please make sure that your seat belt is security fastened. Thank you.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 68 ], "packet_id": 23, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "Ladies and gentlemen, we are now ready for take off. Please make sure that your seat belt is security fastened. Thank you.", "after": "Ladies and gentlemen, we are now ready for take off. Please make sure that your seat belt is security fastened. Thank you.", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard English safety announcement; matches ASR and context perfectly." } ], "language": "en", "packet_ids": [ 23 ], "index": 68, "source_text": "Ladies and gentlemen, we are now ready for take off. Please make sure that your seat belt is security fastened. Thank you.", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Standard English safety announcement; matches ASR and context perfectly.", "raw_word_timestamps": [ { "word": " Ladies", "start": 424.06, "end": 424.58, "probability": 0.732421875 }, { "word": " and", "start": 424.58, "end": 424.84, "probability": 0.9873046875 }, { "word": " gentlemen,", "start": 424.84, "end": 425.26, "probability": 0.8330078125 }, { "word": " we", "start": 425.46, "end": 425.52, "probability": 0.97021484375 }, { "word": " are", "start": 425.52, "end": 425.65999999999997, "probability": 0.97998046875 }, { "word": " now", "start": 425.65999999999997, "end": 425.84, "probability": 0.9765625 }, { "word": " ready", "start": 425.84, "end": 426.14, "probability": 0.99365234375 }, { "word": " for", "start": 426.14, "end": 426.4, "probability": 0.89013671875 }, { "word": " take", "start": 426.4, "end": 426.65999999999997, "probability": 0.90283203125 }, { "word": " off.", "start": 426.65999999999997, "end": 426.97999999999996, "probability": 0.332275390625 }, { "word": " Please", "start": 427.28, "end": 427.79999999999995, "probability": 0.99169921875 }, { "word": " make", "start": 427.79999999999995, "end": 428.06, "probability": 0.9794921875 }, { "word": " sure", "start": 428.06, "end": 428.32, "probability": 0.99853515625 }, { "word": " that", "start": 428.32, "end": 428.56, "probability": 0.9853515625 }, { "word": " your", "start": 428.56, "end": 428.76, "probability": 0.96435546875 }, { "word": " seat", "start": 428.76, "end": 429.09999999999997, "probability": 0.99853515625 }, { "word": " belt", "start": 429.09999999999997, "end": 429.41999999999996, "probability": 0.59375 }, { "word": " is", "start": 429.41999999999996, "end": 429.62, "probability": 0.998046875 }, { "word": " security", "start": 429.62, "end": 430.02, "probability": 0.56201171875 }, { "word": " fastened.", "start": 430.02, "end": 430.71999999999997, "probability": 0.9091796875 }, { "word": " Thank", "start": 430.96, "end": 431.17999999999995, "probability": 0.99462890625 }, { "word": " you.", "start": 431.17999999999995, "end": 431.41999999999996, "probability": 0.9873046875 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 433.5, "end": 438.76, "text": "各位貴賓,飛機即將起飛,請您再次確認是否確實繫好安全帶,謝謝", "raw_asr_text": "各位靈民,飛機即將起飛,請您再次確認是否確實積好安全帶,謝謝", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 24, "words": [], "asr_metrics": { "avg_logprob": -0.5224617719650269, "compression_ratio": 0.9032258064516129, "no_speech_prob": 1.4718051430384627e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "各位靈民,飛機即將起飛,請您再次確認是否確實積好安全帶,謝謝" } ], "id": 69, "review_annotation": "", "row_id": 69, "bounded_row_id": 69, "parent_row_ids": [ 69 ], "parent_row_id": 69, "packet_id": 24, "LID_evidence": [ { "region_id": 55, "probabilities": { "ja": 0.012259923198143952, "zh": 0.9074018858373165, "en": 0.01971785213390831 }, "status": "STABLE" } ], "VAD_region_ids": [ 55 ], "language_confidence": 0.9074018858373165, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 432.7, "packet_end": 440.0, "raw_text": "各位靈民,飛機即將起飛,請您再次確認是否確實積好安全帶,謝謝", "selected_text": "各位貴賓,飛機即將起飛,請您再次確認是否確實繫好安全帶,謝謝", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 69 ], "packet_id": 24, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "各位靈民,飛機即將起飛,請您再次確認是否確實積好安全帶,謝謝", "after": "各位貴賓,飛機即將起飛,請您再次確認是否確實繫好安全帶,謝謝", "classification": "STRONG_RESTORATION", "reason": "ASR '靈民' and '積' are phonetic errors for '貴賓' and '繫', corrected based on context." } ], "language": "zh", "packet_ids": [ 24 ], "index": 69, "independent_ASR": { "text": "各位贵宾，飞机即将起飞，请您再次确认是否确实系好安全带，谢谢。", "chunks": [ { "text": "各位贵宾，飞机即将起飞，请您再次确认是否确实系好安全带，谢谢。", "start": 0.0, "end": 7.3, "chunk_index": 0, "language": "Chinese", "finish_reason": "eos", "truncated": false, "generated_tokens": 19, "max_new_tokens": 128 } ], "detected_language": "Chinese", "segments": [ { "start": 432.7, "end": 440.0, "text": "各位贵宾，飞机即将起飞，请您再次确认是否确实系好安全带，谢谢。" } ], "start": 432.7, "end": 440.0, "engine": "mlx-qwen3-asr", "language": "zh" }, "raw_word_timestamps": [ { "word": "各位", "start": 433.5, "end": 433.86, "probability": 0.947265625 }, { "word": "靈", "start": 433.86, "end": 434.08, "probability": 0.301025390625 }, { "word": "民,", "start": 434.08, "end": 434.24, "probability": 0.342041015625 }, { "word": "飛", "start": 434.38, "end": 434.52, "probability": 0.481689453125 }, { "word": "機", "start": 434.52, "end": 434.65999999999997, "probability": 0.9951171875 }, { "word": "即", "start": 434.65999999999997, "end": 434.78, "probability": 0.998046875 }, { "word": "將", "start": 434.78, "end": 434.96, "probability": 0.99169921875 }, { "word": "起", "start": 434.96, "end": 435.18, "probability": 0.99072265625 }, { "word": "飛,", "start": 435.18, "end": 435.44, "probability": 0.99560546875 }, { "word": "請", "start": 435.84, "end": 435.97999999999996, "probability": 0.9755859375 }, { "word": "您", "start": 435.97999999999996, "end": 436.09999999999997, "probability": 0.279052734375 }, { "word": "再", "start": 436.09999999999997, "end": 436.26, "probability": 0.9853515625 }, { "word": "次", "start": 436.26, "end": 436.47999999999996, "probability": 0.9794921875 }, { "word": "確", "start": 436.47999999999996, "end": 436.65999999999997, "probability": 0.98876953125 }, { "word": "認", "start": 436.65999999999997, "end": 436.82, "probability": 0.9912109375 }, { "word": "是", "start": 436.82, "end": 436.96, "probability": 0.9755859375 }, { "word": "否", "start": 436.96, "end": 437.12, "probability": 0.99951171875 }, { "word": "確", "start": 437.12, "end": 437.3, "probability": 0.99853515625 }, { "word": "實", "start": 437.3, "end": 437.53999999999996, "probability": 0.9990234375 }, { "word": "積", "start": 437.53999999999996, "end": 437.62, "probability": 0.57470703125 }, { "word": "好", "start": 437.62, "end": 437.82, "probability": 0.98046875 }, { "word": "安全", "start": 437.82, "end": 438.15999999999997, "probability": 0.99755859375 }, { "word": "帶,", "start": 438.15999999999997, "end": 438.44, "probability": 0.96875 }, { "word": "謝謝", "start": 438.56, "end": 438.76, "probability": 0.990234375 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "各位貴賓,飛機即將起飛,請您再次確認是否確實繫好安全帶,謝謝", "review_status": "STRONG_RESTORATION", "classification": "STRONG_RESTORATION", "confidence": 0.9, "correction_reason": "ASR '靈民' and '積' are phonetic errors for '貴賓' and '繫', corrected based on context." }, { "start": 442.1, "end": 450.52000000000004, "text": "皆さま、羽田空港到着時刻は3月14日午後5時30分の予定でございます。", "raw_asr_text": "皆さま、羽田空港到着時刻は3月14日午後5時30分の予定でございます。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 25, "words": [], "asr_metrics": { "avg_logprob": -0.12873430813060088, "compression_ratio": 0.93, "no_speech_prob": 5.9811222814787346e-12 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "皆さま、羽田空港到着時刻は3月14日午後5時30分の予定でございます。" } ], "id": 70, "review_annotation": "", "row_id": 70, "bounded_row_id": 70, "parent_row_ids": [ 70 ], "parent_row_id": 70, "packet_id": 25, "LID_evidence": [ { "region_id": 56, "probabilities": { "ja": 0, "zh": 0, "en": 0 }, "status": "UNCERTAIN" }, { "region_id": 57, "probabilities": { "ja": 0.9897016021940443, "zh": 0.0009392024836011438, "en": 0.005060869217332866 }, "status": "STABLE" } ], "VAD_region_ids": [ 56, 57 ], "language_confidence": 0.9897016021940443, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 441.3, "packet_end": 451.7, "raw_text": "皆さま、羽田空港到着時刻は3月14日午後5時30分の予定でございます。", "selected_text": "皆さま、羽田空港到着時刻は3月14日午後5時30分の予定でございます。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 70 ], "packet_id": 25, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "皆さま、羽田空港到着時刻は3月14日午後5時30分の予定でございます。", "after": "皆さま、羽田空港到着時刻は3月14日午後5時30分の予定でございます。", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard announcement phrasing matches context and parallel text perfectly." } ], "language": "ja", "packet_ids": [ 25 ], "index": 70, "source_text": "皆さま、羽田空港到着時刻は3月14日午後5時30分の予定でございます。", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Standard announcement phrasing matches context and parallel text perfectly.", "raw_word_timestamps": [ { "word": "皆", "start": 441.94, "end": 442.42, "probability": 0.81201171875 }, { "word": "さ", "start": 442.42, "end": 442.58, "probability": 0.53857421875 }, { "word": "ま、", "start": 442.58, "end": 442.82, "probability": 0.99951171875 }, { "word": "羽", "start": 443.52000000000004, "end": 444.98, "probability": 0.99609375 }, { "word": "田", "start": 444.98, "end": 445.14, "probability": 0.9912109375 }, { "word": "空", "start": 445.14, "end": 445.3, "probability": 0.99853515625 }, { "word": "港", "start": 445.3, "end": 445.54, "probability": 1.0 }, { "word": "到", "start": 445.54, "end": 445.7, "probability": 0.98828125 }, { "word": "着", "start": 445.7, "end": 445.96000000000004, "probability": 1.0 }, { "word": "時", "start": 445.96000000000004, "end": 446.16, "probability": 0.99853515625 }, { "word": "刻", "start": 446.16, "end": 446.40000000000003, "probability": 0.99951171875 }, { "word": "は", "start": 446.40000000000003, "end": 446.6, "probability": 0.9990234375 }, { "word": "3", "start": 446.6, "end": 446.88, "probability": 0.654296875 }, { "word": "月", "start": 446.88, "end": 447.1, "probability": 0.99951171875 }, { "word": "14", "start": 447.1, "end": 447.58, "probability": 0.9990234375 }, { "word": "日", "start": 447.58, "end": 447.92, "probability": 1.0 }, { "word": "午", "start": 447.92, "end": 448.32, "probability": 0.90380859375 }, { "word": "後", "start": 448.32, "end": 448.52000000000004, "probability": 0.9990234375 }, { "word": "5", "start": 448.52000000000004, "end": 448.68, "probability": 0.9970703125 }, { "word": "時", "start": 448.68, "end": 448.90000000000003, "probability": 1.0 }, { "word": "30", "start": 448.90000000000003, "end": 449.2, "probability": 0.99951171875 }, { "word": "分", "start": 449.2, "end": 449.5, "probability": 1.0 }, { "word": "の", "start": 449.5, "end": 449.62, "probability": 0.99951171875 }, { "word": "予", "start": 449.62, "end": 449.78000000000003, "probability": 0.9990234375 }, { "word": "定", "start": 449.78000000000003, "end": 449.88, "probability": 1.0 }, { "word": "で", "start": 449.88, "end": 450.02000000000004, "probability": 0.99755859375 }, { "word": "ございます。", "start": 450.02000000000004, "end": 450.52000000000004, "probability": 1.0 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 456.3, "end": 465.66, "text": "松山と東京の間には1時間の時差があり、ただいまの東京の時刻は午後2時40分でございます。", "raw_asr_text": "東山と東京の間には1時間の時差があり、ただいまの東京の時刻は午後2時40分でございます。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 26, "words": [], "asr_metrics": { "avg_logprob": -0.21507438860441508, "compression_ratio": 1.1923076923076923, "no_speech_prob": 6.648199352143891e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "東山と東京の間には1時間の時差があり、ただいまの東京の時刻は午後2時40分でございます。" } ], "id": 71, "review_annotation": "", "row_id": 71, "bounded_row_id": 71, "parent_row_ids": [ 71 ], "parent_row_id": 71, "packet_id": 26, "LID_evidence": [ { "region_id": 58, "probabilities": { "ja": 0.9842967711962186, "zh": 0.0013658483176886176, "en": 0.007229176242477619 }, "status": "STABLE" }, { "region_id": 59, "probabilities": { "ja": 0.8788296182950338, "zh": 0.008788938954239711, "en": 0.01106165962604185 }, "status": "STABLE" } ], "VAD_region_ids": [ 58, 59 ], "language_confidence": 0.9842967711962186, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 455.5, "packet_end": 466.8, "raw_text": "東山と東京の間には1時間の時差があり、ただいまの東京の時刻は午後2時40分でございます。", "selected_text": "松山と東京の間には1時間の時差があり、ただいまの東京の時刻は午後2時40分でございます。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 71 ], "packet_id": 26, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "東山と東京の間には1時間の時差があり、ただいまの東京の時刻は午後2時40分でございます。", "after": "松山と東京の間には1時間の時差があり、ただいまの東京の時刻は午後2時40分でございます。", "classification": "STRONG_RESTORATION", "reason": "ASR '東山' is a phonetic error for '松山' (Songshan Airport), supported by flight route." } ], "language": "ja", "packet_ids": [ 26 ], "index": 71, "independent_ASR": { "text": "ソーシャンと東京の間には1時間の時差があり、ただいまの東京の時刻は午後2時40分でございます。", "chunks": [ { "text": "ソーシャンと東京の間には1時間の時差があり、ただいまの東京の時刻は午後2時40分でございます。", "start": 0.0, "end": 11.3, "chunk_index": 0, "language": "Japanese", "finish_reason": "eos", "truncated": false, "generated_tokens": 36, "max_new_tokens": 168 } ], "detected_language": "Japanese", "segments": [ { "start": 455.5, "end": 466.8, "text": "ソーシャンと東京の間には1時間の時差があり、ただいまの東京の時刻は午後2時40分でございます。" } ], "start": 455.5, "end": 466.8, "engine": "mlx-qwen3-asr", "language": "ja" }, "raw_word_timestamps": [ { "word": "東", "start": 456.08, "end": 456.52, "probability": 0.265869140625 }, { "word": "山", "start": 456.52, "end": 456.7, "probability": 0.6884765625 }, { "word": "と", "start": 456.7, "end": 456.88, "probability": 0.9140625 }, { "word": "東", "start": 456.88, "end": 457.1, "probability": 0.99853515625 }, { "word": "京", "start": 457.1, "end": 457.26, "probability": 0.99951171875 }, { "word": "の", "start": 457.26, "end": 457.42, "probability": 0.9970703125 }, { "word": "間", "start": 457.42, "end": 457.56, "probability": 0.9990234375 }, { "word": "には", "start": 457.56, "end": 457.94, "probability": 0.99267578125 }, { "word": "1", "start": 457.94, "end": 458.26, "probability": 0.8662109375 }, { "word": "時間", "start": 458.26, "end": 458.58, "probability": 0.9990234375 }, { "word": "の", "start": 458.58, "end": 458.84, "probability": 0.9970703125 }, { "word": "時", "start": 458.84, "end": 458.94, "probability": 0.99951171875 }, { "word": "差", "start": 458.94, "end": 459.08, "probability": 0.998046875 }, { "word": "が", "start": 459.08, "end": 459.24, "probability": 0.9970703125 }, { "word": "あり、", "start": 459.24, "end": 459.52, "probability": 0.99853515625 }, { "word": "ただ", "start": 459.92, "end": 460.34, "probability": 0.8671875 }, { "word": "い", "start": 460.34, "end": 460.5, "probability": 0.95068359375 }, { "word": "ま", "start": 460.5, "end": 460.66, "probability": 0.99951171875 }, { "word": "の", "start": 460.66, "end": 460.9, "probability": 0.99609375 }, { "word": "東", "start": 460.9, "end": 461.2, "probability": 0.99853515625 }, { "word": "京", "start": 461.2, "end": 461.34, "probability": 1.0 }, { "word": "の", "start": 461.34, "end": 461.5, "probability": 0.9990234375 }, { "word": "時", "start": 461.5, "end": 461.64, "probability": 0.99951171875 }, { "word": "刻", "start": 461.64, "end": 461.88, "probability": 0.99951171875 }, { "word": "は", "start": 461.88, "end": 462.18, "probability": 0.99951171875 }, { "word": "午", "start": 462.18, "end": 463.42, "probability": 0.94921875 }, { "word": "後", "start": 463.42, "end": 463.6, "probability": 0.9990234375 }, { "word": "2", "start": 463.6, "end": 463.78, "probability": 0.998046875 }, { "word": "時", "start": 463.78, "end": 464.0, "probability": 0.99951171875 }, { "word": "40", "start": 464.0, "end": 464.72, "probability": 0.99755859375 }, { "word": "分", "start": 464.72, "end": 465.04, "probability": 1.0 }, { "word": "で", "start": 465.04, "end": 465.16, "probability": 0.99609375 }, { "word": "ございます。", "start": 465.16, "end": 465.66, "probability": 1.0 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "松山と東京の間には1時間の時差があり、ただいまの東京の時刻は午後2時40分でございます。", "review_status": "STRONG_RESTORATION", "classification": "STRONG_RESTORATION", "confidence": 0.9, "correction_reason": "ASR '東山' is a phonetic error for '松山' (Songshan Airport), supported by flight route." }, { "start": 471.1, "end": 475.2, "text": "毛布をご使用の際は、毛布の上からシートベルトをお締めください。", "raw_asr_text": "毛布をご使用の際は、毛布の上からシートベルトをお締めください。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 27, "words": [], "asr_metrics": { "avg_logprob": -0.1571191151936849, "compression_ratio": 1.0568181818181819, "no_speech_prob": 5.6640712085831435e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "毛布をご使用の際は、毛布の上からシートベルトをお締めください。" } ], "id": 72, "review_annotation": "", "row_id": 72, "bounded_row_id": 72, "parent_row_ids": [ 72 ], "parent_row_id": 72, "packet_id": 27, "LID_evidence": [ { "region_id": 60, "probabilities": { "ja": 0.9835621317227682, "zh": 0.0010900165968147728, "en": 0.008110727304786755 }, "status": "STABLE" } ], "VAD_region_ids": [ 60 ], "language_confidence": 0.9835621317227682, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 470.3, "packet_end": 476.6, "raw_text": "毛布をご使用の際は、毛布の上からシートベルトをお締めください。", "selected_text": "毛布をご使用の際は、毛布の上からシートベルトをお締めください。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 72 ], "packet_id": 27, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "毛布をご使用の際は、毛布の上からシートベルトをお締めください。", "after": "毛布をご使用の際は、毛布の上からシートベルトをお締めください。", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard safety instruction, grammatically correct and contextually appropriate." } ], "language": "ja", "packet_ids": [ 27 ], "index": 72, "source_text": "毛布をご使用の際は、毛布の上からシートベルトをお締めください。", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Standard safety instruction, grammatically correct and contextually appropriate.", "raw_word_timestamps": [ { "word": "毛", "start": 471.08, "end": 471.40000000000003, "probability": 0.97216796875 }, { "word": "布", "start": 471.40000000000003, "end": 471.56, "probability": 0.99853515625 }, { "word": "を", "start": 471.56, "end": 471.68, "probability": 0.9951171875 }, { "word": "ご", "start": 471.68, "end": 471.76, "probability": 0.9814453125 }, { "word": "使", "start": 471.76, "end": 471.92, "probability": 0.99755859375 }, { "word": "用", "start": 471.92, "end": 472.04, "probability": 0.99951171875 }, { "word": "の", "start": 472.04, "end": 472.14, "probability": 0.9775390625 }, { "word": "際", "start": 472.14, "end": 472.38, "probability": 0.99951171875 }, { "word": "は、", "start": 472.38, "end": 472.62, "probability": 0.99462890625 }, { "word": "毛", "start": 472.7, "end": 472.86, "probability": 0.96630859375 }, { "word": "布", "start": 472.86, "end": 473.02000000000004, "probability": 1.0 }, { "word": "の", "start": 473.02000000000004, "end": 473.12, "probability": 0.99951171875 }, { "word": "上", "start": 473.12, "end": 473.3, "probability": 0.99951171875 }, { "word": "から", "start": 473.3, "end": 473.56, "probability": 1.0 }, { "word": "シ", "start": 473.56, "end": 473.76, "probability": 0.99560546875 }, { "word": "ート", "start": 473.76, "end": 473.92, "probability": 0.9580078125 }, { "word": "ベ", "start": 473.92, "end": 474.08, "probability": 0.9990234375 }, { "word": "ル", "start": 474.08, "end": 474.18, "probability": 0.99951171875 }, { "word": "ト", "start": 474.18, "end": 474.3, "probability": 0.99951171875 }, { "word": "を", "start": 474.3, "end": 474.48, "probability": 0.99853515625 }, { "word": "お", "start": 474.48, "end": 474.6, "probability": 0.947265625 }, { "word": "締", "start": 474.6, "end": 474.72, "probability": 0.697265625 }, { "word": "め", "start": 474.72, "end": 474.8, "probability": 0.99951171875 }, { "word": "ください。", "start": 474.8, "end": 475.2, "probability": 0.9833984375 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 479.1, "end": 482.68, "text": "液体類の機内持ち込み制限についてご案内いたします。", "raw_asr_text": "液体類の機内持ち込み制限についてご案内いたします。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 28, "words": [], "asr_metrics": { "avg_logprob": -0.10822569240223277, "compression_ratio": 1.3568464730290457, "no_speech_prob": 5.777246302879346e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "液体類の機内持ち込み制限についてご案内いたします。出国までに購入された1置きあたり100mlを超える液体、ジェル、クリーム、エアゾール類は、国際線への乗り継ぎ便及びお帰りの便の客室への持ち込みはできませんので、ご注意ください。" } ], "id": 73, "review_annotation": "", "row_id": 73, "bounded_row_id": 73, "parent_row_ids": [ 73 ], "parent_row_id": 73, "packet_id": 28, "LID_evidence": [ { "region_id": 61, "probabilities": { "ja": 0.9891727805137634, "zh": 0.0008040153566980735, "en": 0.005621545028407127 }, "status": "STABLE" } ], "VAD_region_ids": [ 61 ], "language_confidence": 0.9891727805137634, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 478.3, "packet_end": 501.0, "raw_text": "液体類の機内持ち込み制限についてご案内いたします。", "selected_text": "液体類の機内持ち込み制限についてご案内いたします。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 73 ], "packet_id": 28, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "液体類の機内持ち込み制限についてご案内いたします。", "after": "液体類の機内持ち込み制限についてご案内いたします。", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard introductory phrase for liquid restrictions, matches context." } ], "language": "ja", "packet_ids": [ 28 ], "index": 73, "source_text": "液体類の機内持ち込み制限についてご案内いたします。", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Standard introductory phrase for liquid restrictions, matches context.", "raw_word_timestamps": [ { "word": "液", "start": 479.04, "end": 479.40000000000003, "probability": 0.9970703125 }, { "word": "体", "start": 479.40000000000003, "end": 479.56, "probability": 0.99951171875 }, { "word": "類", "start": 479.56, "end": 479.72, "probability": 0.99951171875 }, { "word": "の", "start": 479.72, "end": 479.92, "probability": 1.0 }, { "word": "機", "start": 479.92, "end": 480.12, "probability": 0.96923828125 }, { "word": "内", "start": 480.12, "end": 480.26, "probability": 0.99853515625 }, { "word": "持", "start": 480.26, "end": 480.46000000000004, "probability": 0.9990234375 }, { "word": "ち", "start": 480.46000000000004, "end": 480.56, "probability": 0.984375 }, { "word": "込", "start": 480.56, "end": 480.66, "probability": 1.0 }, { "word": "み", "start": 480.66, "end": 480.82, "probability": 1.0 }, { "word": "制", "start": 480.82, "end": 480.92, "probability": 0.9990234375 }, { "word": "限", "start": 480.92, "end": 481.12, "probability": 0.99951171875 }, { "word": "に", "start": 481.12, "end": 481.26, "probability": 0.9990234375 }, { "word": "つ", "start": 481.26, "end": 481.42, "probability": 1.0 }, { "word": "いて", "start": 481.42, "end": 481.64, "probability": 1.0 }, { "word": "ご", "start": 481.64, "end": 481.84000000000003, "probability": 0.9716796875 }, { "word": "案", "start": 481.84000000000003, "end": 482.02000000000004, "probability": 0.9990234375 }, { "word": "内", "start": 482.02000000000004, "end": 482.12, "probability": 0.99951171875 }, { "word": "いた", "start": 482.12, "end": 482.3, "probability": 0.99755859375 }, { "word": "します。", "start": 482.3, "end": 482.68, "probability": 1.0 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 484.22, "end": 492.74, "text": "出国までに購入された一容器当たり100mlを超える液体、ジェル、クリーム、エアゾール類は、", "raw_asr_text": "出国までに購入された1置きあたり100mlを超える液体、ジェル、クリーム、エアゾール類は、", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 28, "words": [], "asr_metrics": { "avg_logprob": -0.10822569240223277, "compression_ratio": 1.3568464730290457, "no_speech_prob": 5.777246302879346e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "液体類の機内持ち込み制限についてご案内いたします。出国までに購入された1置きあたり100mlを超える液体、ジェル、クリーム、エアゾール類は、国際線への乗り継ぎ便及びお帰りの便の客室への持ち込みはできませんので、ご注意ください。" } ], "id": 74, "review_annotation": "", "row_id": 74, "bounded_row_id": 74, "parent_row_ids": [ 74 ], "parent_row_id": 74, "packet_id": 28, "LID_evidence": [ { "region_id": 62, "probabilities": { "ja": 0.9811633030573527, "zh": 0.0010029873163956734, "en": 0.009657866722490225 }, "status": "STABLE" } ], "VAD_region_ids": [ 62 ], "language_confidence": 0.9811633030573527, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 478.3, "packet_end": 501.0, "raw_text": "出国までに購入された1置きあたり100mlを超える液体、ジェル、クリーム、エアゾール類は、", "selected_text": "出国までに購入された一容器当たり100mlを超える液体、ジェル、クリーム、エアゾール類は、", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 74 ], "packet_id": 28, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "出国までに購入された1置きあたり100mlを超える液体、ジェル、クリーム、エアゾール類は、", "after": "出国までに購入された一容器当たり100mlを超える液体、ジェル、クリーム、エアゾール類は、", "classification": "STRONG_RESTORATION", "reason": "ASR '1置きあたり' is a phonetic error for '一容器当たり' (per container), standard regulation." } ], "language": "ja", "packet_ids": [ 28 ], "index": 74, "independent_ASR": { "text": "液体類の機内持ち込み制限についてご案内いたします。出国までに購入された一容器当たり100ミリリットルを超える液体、ジェル、クリーム、エアゾル類は、国際線への乗り継ぎ便およびお帰りの便の客室での持ち込みはできませんのでご注意ください。", "chunks": [ { "text": "液体類の機内持ち込み制限についてご案内いたします。出国までに購入された一容器当たり100ミリリットルを超える液体、ジェル、クリーム、エアゾル類は、国際線への乗り継ぎ便およびお帰りの便の客室での持ち込みはできませんのでご注意ください。", "start": 0.0, "end": 22.7, "chunk_index": 0, "language": "Japanese", "finish_reason": "eos", "truncated": false, "generated_tokens": 75, "max_new_tokens": 305 } ], "detected_language": "Japanese", "segments": [ { "start": 478.3, "end": 501.0, "text": "液体類の機内持ち込み制限についてご案内いたします。出国までに購入された一容器当たり100ミリリットルを超える液体、ジェル、クリーム、エアゾル類は、国際線への乗り継ぎ便およびお帰りの便の客室での持ち込みはできませんのでご注意ください。" } ], "start": 478.3, "end": 501.0, "engine": "mlx-qwen3-asr", "language": "ja" }, "raw_word_timestamps": [ { "word": "出", "start": 484.22, "end": 484.58, "probability": 0.9970703125 }, { "word": "国", "start": 484.58, "end": 484.76, "probability": 0.9931640625 }, { "word": "まで", "start": 484.76, "end": 485.06, "probability": 1.0 }, { "word": "に", "start": 485.06, "end": 485.26, "probability": 1.0 }, { "word": "購", "start": 485.26, "end": 485.56, "probability": 1.0 }, { "word": "入", "start": 485.56, "end": 485.7, "probability": 1.0 }, { "word": "さ", "start": 485.7, "end": 485.90000000000003, "probability": 1.0 }, { "word": "れた", "start": 485.90000000000003, "end": 486.16, "probability": 1.0 }, { "word": "1", "start": 486.16, "end": 486.56, "probability": 0.931640625 }, { "word": "置", "start": 486.56, "end": 486.7, "probability": 0.27587890625 }, { "word": "き", "start": 486.7, "end": 486.88, "probability": 0.984375 }, { "word": "あ", "start": 486.88, "end": 487.0, "probability": 0.81591796875 }, { "word": "た", "start": 487.0, "end": 487.12, "probability": 1.0 }, { "word": "り", "start": 487.12, "end": 487.26, "probability": 1.0 }, { "word": "100", "start": 487.26, "end": 487.58, "probability": 0.99658203125 }, { "word": "ml", "start": 487.58, "end": 487.88, "probability": 0.93994140625 }, { "word": "を", "start": 487.88, "end": 488.38, "probability": 0.99755859375 }, { "word": "超", "start": 488.38, "end": 488.56, "probability": 1.0 }, { "word": "える", "start": 488.56, "end": 488.82, "probability": 1.0 }, { "word": "液", "start": 488.82, "end": 489.54, "probability": 0.97412109375 }, { "word": "体、", "start": 489.54, "end": 489.8, "probability": 1.0 }, { "word": "ジ", "start": 489.98, "end": 490.24, "probability": 0.9921875 }, { "word": "ェ", "start": 490.24, "end": 490.38, "probability": 1.0 }, { "word": "ル、", "start": 490.38, "end": 490.42, "probability": 0.9990234375 }, { "word": "ク", "start": 490.82, "end": 491.02000000000004, "probability": 0.56396484375 }, { "word": "リ", "start": 491.02000000000004, "end": 491.1, "probability": 0.998046875 }, { "word": "ーム、", "start": 491.1, "end": 491.34000000000003, "probability": 0.99951171875 }, { "word": "エ", "start": 491.72, "end": 491.88, "probability": 0.99853515625 }, { "word": "ア", "start": 491.88, "end": 491.98, "probability": 0.998046875 }, { "word": "ゾ", "start": 491.98, "end": 492.14, "probability": 0.998046875 }, { "word": "ール", "start": 492.14, "end": 492.28000000000003, "probability": 0.87744140625 }, { "word": "類", "start": 492.28000000000003, "end": 492.46000000000004, "probability": 1.0 }, { "word": "は、", "start": 492.46000000000004, "end": 492.74, "probability": 1.0 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "出国までに購入された一容器当たり100mlを超える液体、ジェル、クリーム、エアゾール類は、", "review_status": "STRONG_RESTORATION", "classification": "STRONG_RESTORATION", "confidence": 0.9, "correction_reason": "ASR '1置きあたり' is a phonetic error for '一容器当たり' (per container), standard regulation." }, { "start": 493.56, "end": 499.56, "text": "国際線への乗り継ぎ便及びお帰りの便の客室への持ち込みはできませんので、ご注意ください。", "raw_asr_text": "国際線への乗り継ぎ便及びお帰りの便の客室への持ち込みはできませんので、ご注意ください。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 28, "words": [ { "word": "国", "start": 493.56, "end": 493.86, "probability": 0.99462890625 }, { "word": "際", "start": 493.86, "end": 494.02000000000004, "probability": 0.99853515625 }, { "word": "線", "start": 494.02000000000004, "end": 494.26, "probability": 0.98193359375 }, { "word": "へ", "start": 494.26, "end": 494.42, "probability": 0.99951171875 }, { "word": "の", "start": 494.42, "end": 494.56, "probability": 1.0 }, { "word": "乗", "start": 494.56, "end": 494.7, "probability": 0.99462890625 }, { "word": "り", "start": 494.7, "end": 494.82, "probability": 0.9931640625 }, { "word": "継", "start": 494.82, "end": 494.96000000000004, "probability": 0.998046875 }, { "word": "ぎ", "start": 494.96000000000004, "end": 495.08000000000004, "probability": 0.998046875 }, { "word": "便", "start": 495.08000000000004, "end": 495.26, "probability": 0.9482421875 }, { "word": "及", "start": 495.26, "end": 495.64, "probability": 0.461181640625 }, { "word": "び", "start": 495.64, "end": 495.88, "probability": 1.0 }, { "word": "お", "start": 495.88, "end": 496.02, "probability": 0.98681640625 }, { "word": "帰", "start": 496.02, "end": 496.18, "probability": 0.9521484375 }, { "word": "り", "start": 496.18, "end": 496.28000000000003, "probability": 1.0 }, { "word": "の", "start": 496.28000000000003, "end": 496.38, "probability": 1.0 }, { "word": "便", "start": 496.38, "end": 496.52, "probability": 0.9990234375 }, { "word": "の", "start": 496.52, "end": 496.68, "probability": 0.99951171875 }, { "word": "客", "start": 496.68, "end": 496.86, "probability": 0.99755859375 }, { "word": "室", "start": 496.86, "end": 497.1, "probability": 0.9990234375 }, { "word": "へ", "start": 497.1, "end": 497.22, "probability": 0.9990234375 }, { "word": "の", "start": 497.22, "end": 497.32, "probability": 1.0 }, { "word": "持", "start": 497.32, "end": 497.5, "probability": 0.9990234375 }, { "word": "ち", "start": 497.5, "end": 497.6, "probability": 1.0 }, { "word": "込", "start": 497.6, "end": 497.68, "probability": 1.0 }, { "word": "み", "start": 497.68, "end": 497.78000000000003, "probability": 1.0 }, { "word": "は", "start": 497.78000000000003, "end": 497.90000000000003, "probability": 0.9990234375 }, { "word": "でき", "start": 497.90000000000003, "end": 498.08000000000004, "probability": 0.99951171875 }, { "word": "ません", "start": 498.08000000000004, "end": 498.40000000000003, "probability": 0.99951171875 }, { "word": "ので、", "start": 498.40000000000003, "end": 498.74, "probability": 0.99951171875 }, { "word": "ご", "start": 498.88, "end": 498.96000000000004, "probability": 0.99853515625 }, { "word": "注意", "start": 498.96000000000004, "end": 499.18, "probability": 1.0 }, { "word": "ください。", "start": 499.18, "end": 499.56, "probability": 0.99951171875 } ], "asr_metrics": { "avg_logprob": -0.10822569240223277, "compression_ratio": 1.3568464730290457, "no_speech_prob": 5.777246302879346e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "液体類の機内持ち込み制限についてご案内いたします。出国までに購入された1置きあたり100mlを超える液体、ジェル、クリーム、エアゾール類は、国際線への乗り継ぎ便及びお帰りの便の客室への持ち込みはできませんので、ご注意ください。" } ], "id": 75, "review_annotation": "", "row_id": 75, "bounded_row_id": 75, "parent_row_ids": [ 75 ], "parent_row_id": 75, "packet_id": 28, "LID_evidence": [ { "region_id": 63, "probabilities": { "ja": 0.9926614207880837, "zh": 0.0005143823675877814, "en": 0.003927687316068581 }, "status": "STABLE" } ], "VAD_region_ids": [ 63 ], "language_confidence": 0.9926614207880837, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 478.3, "packet_end": 501.0, "raw_text": "国際線への乗り継ぎ便及びお帰りの便の客室への持ち込みはできませんので、ご注意ください。", "selected_text": "国際線への乗り継ぎ便及びお帰りの便の客室への持ち込みはできませんので、ご注意ください。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 75 ], "packet_id": 28, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "国際線への乗り継ぎ便及びお帰りの便の客室への持ち込みはできませんので、ご注意ください。", "after": "国際線への乗り継ぎ便及びお帰りの便の客室への持ち込みはできませんので、ご注意ください。", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard regulatory warning, grammatically correct and matches context." } ], "language": "ja", "packet_ids": [ 28 ], "index": 75, "source_text": "国際線への乗り継ぎ便及びお帰りの便の客室への持ち込みはできませんので、ご注意ください。", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Standard regulatory warning, grammatically correct and matches context." }, { "start": 502.56, "end": 507.36, "text": "発熱・腹痛などの症状があるお客様は、客室乗務員にお知らせください。", "raw_asr_text": "発熱・腹痛などの症状があるお客様は、客室乗務員にお知らせください。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 29, "words": [ { "word": "発", "start": 502.56, "end": 502.84, "probability": 0.98779296875 }, { "word": "熱", "start": 502.84, "end": 503.12, "probability": 1.0 }, { "word": "・", "start": 503.12, "end": 503.28, "probability": 0.61083984375 }, { "word": "腹", "start": 503.28, "end": 503.44, "probability": 0.9931640625 }, { "word": "痛", "start": 503.44, "end": 503.59999999999997, "probability": 0.99951171875 }, { "word": "な", "start": 503.59999999999997, "end": 503.76, "probability": 0.99853515625 }, { "word": "ど", "start": 503.76, "end": 503.86, "probability": 1.0 }, { "word": "の", "start": 503.86, "end": 503.97999999999996, "probability": 1.0 }, { "word": "症", "start": 503.97999999999996, "end": 504.18, "probability": 1.0 }, { "word": "状", "start": 504.18, "end": 504.34, "probability": 1.0 }, { "word": "が", "start": 504.34, "end": 504.46, "probability": 1.0 }, { "word": "ある", "start": 504.46, "end": 504.58, "probability": 1.0 }, { "word": "お", "start": 504.58, "end": 504.71999999999997, "probability": 0.9990234375 }, { "word": "客", "start": 504.71999999999997, "end": 504.86, "probability": 1.0 }, { "word": "様", "start": 504.86, "end": 505.15999999999997, "probability": 0.9951171875 }, { "word": "は、", "start": 505.15999999999997, "end": 505.38, "probability": 0.99951171875 }, { "word": "客", "start": 505.44, "end": 505.62, "probability": 0.693359375 }, { "word": "室", "start": 505.62, "end": 505.86, "probability": 0.99853515625 }, { "word": "乗", "start": 505.86, "end": 506.03999999999996, "probability": 0.96875 }, { "word": "務", "start": 506.03999999999996, "end": 506.21999999999997, "probability": 0.99951171875 }, { "word": "員", "start": 506.21999999999997, "end": 506.3, "probability": 0.9130859375 }, { "word": "に", "start": 506.3, "end": 506.47999999999996, "probability": 0.99951171875 }, { "word": "お", "start": 506.47999999999996, "end": 506.62, "probability": 1.0 }, { "word": "知", "start": 506.62, "end": 506.74, "probability": 1.0 }, { "word": "ら", "start": 506.74, "end": 506.82, "probability": 1.0 }, { "word": "せ", "start": 506.82, "end": 506.94, "probability": 1.0 }, { "word": "ください。", "start": 506.94, "end": 507.36, "probability": 0.99951171875 } ], "asr_metrics": { "avg_logprob": -0.11505424275117762, "compression_ratio": 1.3333333333333333, "no_speech_prob": 7.078115871195223e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "発熱・腹痛などの症状があるお客様は、客室乗務員にお知らせください。ベルト着用サインは消えましたが、急な揺れに備え、お座りの間は常にシートベルトをお締めください。" } ], "id": 76, "review_annotation": "", "row_id": 76, "bounded_row_id": 76, "parent_row_ids": [ 76 ], "parent_row_id": 76, "packet_id": 29, "LID_evidence": [ { "region_id": 64, "probabilities": { "ja": 0.9885386750102043, "zh": 0.0008993327937787399, "en": 0.005901259937672876 }, "status": "STABLE" } ], "VAD_region_ids": [ 64 ], "language_confidence": 0.9885386750102043, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 501.7, "packet_end": 516.9, "raw_text": "発熱・腹痛などの症状があるお客様は、客室乗務員にお知らせください。", "selected_text": "発熱・腹痛などの症状があるお客様は、客室乗務員にお知らせください。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 76 ], "packet_id": 29, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "発熱・腹痛などの症状があるお客様は、客室乗務員にお知らせください。", "after": "発熱・腹痛などの症状があるお客様は、客室乗務員にお知らせください。", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard health advisory, grammatically correct and matches context." } ], "language": "ja", "packet_ids": [ 29 ], "index": 76, "source_text": "発熱・腹痛などの症状があるお客様は、客室乗務員にお知らせください。", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Standard health advisory, grammatically correct and matches context." }, { "start": 509.38, "end": 512.54, "text": "ベルト着用サインは消えましたが、急な揺れに備え、", "raw_asr_text": "ベルト着用サインは消えましたが、急な揺れに備え、", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 29, "words": [ { "word": "ベ", "start": 509.38, "end": 509.52, "probability": 0.99267578125 }, { "word": "ル", "start": 509.52, "end": 509.59999999999997, "probability": 0.99951171875 }, { "word": "ト", "start": 509.59999999999997, "end": 509.7, "probability": 0.99951171875 }, { "word": "着", "start": 509.7, "end": 509.86, "probability": 0.99951171875 }, { "word": "用", "start": 509.86, "end": 510.09999999999997, "probability": 0.99951171875 }, { "word": "サ", "start": 510.09999999999997, "end": 510.3, "probability": 0.662109375 }, { "word": "イ", "start": 510.3, "end": 510.38, "probability": 1.0 }, { "word": "ン", "start": 510.38, "end": 510.46, "probability": 0.994140625 }, { "word": "は", "start": 510.46, "end": 510.53999999999996, "probability": 0.99951171875 }, { "word": "消", "start": 510.53999999999996, "end": 510.68, "probability": 0.9990234375 }, { "word": "え", "start": 510.68, "end": 510.82, "probability": 1.0 }, { "word": "ました", "start": 510.82, "end": 511.08, "probability": 1.0 }, { "word": "が、", "start": 511.08, "end": 511.32, "probability": 1.0 }, { "word": "急", "start": 511.46, "end": 511.62, "probability": 0.9970703125 }, { "word": "な", "start": 511.62, "end": 511.8, "probability": 0.99951171875 }, { "word": "揺", "start": 511.8, "end": 511.9, "probability": 0.9990234375 }, { "word": "れ", "start": 511.9, "end": 511.97999999999996, "probability": 0.9990234375 }, { "word": "に", "start": 511.97999999999996, "end": 512.16, "probability": 0.9951171875 }, { "word": "備", "start": 512.16, "end": 512.26, "probability": 0.99169921875 }, { "word": "え、", "start": 512.26, "end": 512.54, "probability": 1.0 } ], "asr_metrics": { "avg_logprob": -0.11505424275117762, "compression_ratio": 1.3333333333333333, "no_speech_prob": 7.078115871195223e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "発熱・腹痛などの症状があるお客様は、客室乗務員にお知らせください。ベルト着用サインは消えましたが、急な揺れに備え、お座りの間は常にシートベルトをお締めください。" } ], "id": 77, "review_annotation": "", "row_id": 77, "bounded_row_id": 77, "parent_row_ids": [ 77 ], "parent_row_id": 77, "packet_id": 29, "LID_evidence": [ { "region_id": 65, "probabilities": { "ja": 0.8784256908942812, "zh": 0.001937987534412449, "en": 0.10197309839505968 }, "status": "STABLE" } ], "VAD_region_ids": [ 65 ], "language_confidence": 0.8784256908942812, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 501.7, "packet_end": 516.9, "raw_text": "ベルト着用サインは消えましたが、急な揺れに備え、", "selected_text": "ベルト着用サインは消えましたが、急な揺れに備え、", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 77 ], "packet_id": 29, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "ベルト着用サインは消えましたが、急な揺れに備え、", "after": "ベルト着用サインは消えましたが、急な揺れに備え、", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard safety instruction fragment, grammatically correct." } ], "language": "ja", "packet_ids": [ 29 ], "index": 77, "source_text": "ベルト着用サインは消えましたが、急な揺れに備え、", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Standard safety instruction fragment, grammatically correct." }, { "start": 512.64, "end": 515.54, "text": "お座りの間は常にシートベルトをお締めください。", "raw_asr_text": "お座りの間は常にシートベルトをお締めください。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 29, "words": [ { "word": "お", "start": 512.64, "end": 512.78, "probability": 0.9931640625 }, { "word": "座", "start": 512.78, "end": 512.92, "probability": 0.857421875 }, { "word": "り", "start": 512.92, "end": 513.04, "probability": 1.0 }, { "word": "の", "start": 513.04, "end": 513.16, "probability": 1.0 }, { "word": "間", "start": 513.16, "end": 513.28, "probability": 1.0 }, { "word": "は", "start": 513.28, "end": 513.58, "probability": 1.0 }, { "word": "常", "start": 513.58, "end": 513.8199999999999, "probability": 0.97119140625 }, { "word": "に", "start": 513.8199999999999, "end": 513.96, "probability": 1.0 }, { "word": "シ", "start": 513.96, "end": 514.14, "probability": 0.99755859375 }, { "word": "ート", "start": 514.14, "end": 514.28, "probability": 1.0 }, { "word": "ベ", "start": 514.28, "end": 514.4399999999999, "probability": 0.99951171875 }, { "word": "ル", "start": 514.4399999999999, "end": 514.54, "probability": 1.0 }, { "word": "ト", "start": 514.54, "end": 514.64, "probability": 1.0 }, { "word": "を", "start": 514.64, "end": 514.8199999999999, "probability": 0.97021484375 }, { "word": "お", "start": 514.8199999999999, "end": 514.96, "probability": 0.9912109375 }, { "word": "締", "start": 514.96, "end": 515.02, "probability": 0.9794921875 }, { "word": "め", "start": 515.02, "end": 515.14, "probability": 1.0 }, { "word": "ください。", "start": 515.14, "end": 515.54, "probability": 0.99951171875 } ], "asr_metrics": { "avg_logprob": -0.11505424275117762, "compression_ratio": 1.3333333333333333, "no_speech_prob": 7.078115871195223e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "発熱・腹痛などの症状があるお客様は、客室乗務員にお知らせください。ベルト着用サインは消えましたが、急な揺れに備え、お座りの間は常にシートベルトをお締めください。" } ], "id": 78, "review_annotation": "", "row_id": 78, "bounded_row_id": 78, "parent_row_ids": [ 78 ], "parent_row_id": 78, "packet_id": 29, "LID_evidence": [ { "region_id": 65, "probabilities": { "ja": 0.8784256908942812, "zh": 0.001937987534412449, "en": 0.10197309839505968 }, "status": "STABLE" } ], "VAD_region_ids": [ 65 ], "language_confidence": 0.8784256908942812, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 501.7, "packet_end": 516.9, "raw_text": "お座りの間は常にシートベルトをお締めください。", "selected_text": "お座りの間は常にシートベルトをお締めください。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 78 ], "packet_id": 29, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "お座りの間は常にシートベルトをお締めください。", "after": "お座りの間は常にシートベルトをお締めください。", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard safety instruction, grammatically correct." } ], "language": "ja", "packet_ids": [ 29 ], "index": 78, "source_text": "お座りの間は常にシートベルトをお締めください。", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Standard safety instruction, grammatically correct." }, { "start": 516.7, "end": 523.96, "text": "Ladies and gentlemen, the captain has turned off the seatbelt sign. However, please keep your seatbelt fastened whenever you are seated.", "raw_asr_text": "Ladies and gentlemen, the captain has turned off the seatbelt sign. However, please keep your seatbelt fasten whenever you are seated.", "source_language": "en", "source_lang": "en", "asr_packet_id": 30, "words": [], "asr_metrics": { "avg_logprob": -0.26871026357014977, "compression_ratio": 1.2884615384615385, "no_speech_prob": 4.801061284642749e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "en", "text": "Ladies and gentlemen, the captain has turned off the seatbelt sign. However, please keep your seatbelt fasten whenever you are seated." } ], "id": 79, "review_annotation": "", "row_id": 79, "bounded_row_id": 79, "parent_row_ids": [ 79 ], "parent_row_id": 79, "packet_id": 30, "LID_evidence": [ { "region_id": 66, "probabilities": { "ja": 0.1071992301294813, "zh": 0.009640150925406488, "en": 0.8138890480695409 }, "status": "STABLE" } ], "VAD_region_ids": [ 66 ], "language_confidence": 0.8138890480695409, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 515.9000000000001, "packet_end": 525.3, "raw_text": "Ladies and gentlemen, the captain has turned off the seatbelt sign. However, please keep your seatbelt fasten whenever you are seated.", "selected_text": "Ladies and gentlemen, the captain has turned off the seatbelt sign. However, please keep your seatbelt fastened whenever you are seated.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 79 ], "packet_id": 30, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "Ladies and gentlemen, the captain has turned off the seatbelt sign. However, please keep your seatbelt fasten whenever you are seated.", "after": "Ladies and gentlemen, the captain has turned off the seatbelt sign. However, please keep your seatbelt fastened whenever you are seated.", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "ASR 'fasten' corrected to 'fastened' for grammatical correctness in standard announcement." } ], "language": "en", "packet_ids": [ 30 ], "index": 79, "independent_ASR": { "text": "Ladies and gentlemen, the captain has turned off the seatbelt sign. However, please keep your seatbelt fastened whenever you are seated.", "chunks": [ { "text": "Ladies and gentlemen, the captain has turned off the seatbelt sign. However, please keep your seatbelt fastened whenever you are seated.", "start": 0.0, "end": 9.4, "chunk_index": 0, "language": "English", "finish_reason": "eos", "truncated": false, "generated_tokens": 29, "max_new_tokens": 145 } ], "detected_language": "English", "segments": [ { "start": 515.9000000000001, "end": 525.3000000000001, "text": "Ladies and gentlemen, the captain has turned off the seatbelt sign. However, please keep your seatbelt fastened whenever you are seated." } ], "start": 515.9000000000001, "end": 525.3, "engine": "mlx-qwen3-asr", "language": "en" }, "raw_word_timestamps": [ { "word": " Ladies", "start": 515.9000000000001, "end": 517.0600000000001, "probability": 0.77587890625 }, { "word": " and", "start": 517.0600000000001, "end": 517.3800000000001, "probability": 0.99169921875 }, { "word": " gentlemen,", "start": 517.3800000000001, "end": 517.6600000000001, "probability": 0.8466796875 }, { "word": " the", "start": 517.96, "end": 518.0000000000001, "probability": 0.8994140625 }, { "word": " captain", "start": 518.0000000000001, "end": 518.2200000000001, "probability": 0.86865234375 }, { "word": " has", "start": 518.2200000000001, "end": 518.58, "probability": 0.99462890625 }, { "word": " turned", "start": 518.58, "end": 518.8400000000001, "probability": 0.9970703125 }, { "word": " off", "start": 518.8400000000001, "end": 519.1200000000001, "probability": 0.98828125 }, { "word": " the", "start": 519.1200000000001, "end": 519.32, "probability": 0.986328125 }, { "word": " seatbelt", "start": 519.32, "end": 519.7400000000001, "probability": 0.9033203125 }, { "word": " sign.", "start": 519.7400000000001, "end": 520.1600000000001, "probability": 0.95458984375 }, { "word": " However,", "start": 520.4000000000001, "end": 521.0000000000001, "probability": 0.9931640625 }, { "word": " please", "start": 521.2800000000001, "end": 521.46, "probability": 0.994140625 }, { "word": " keep", "start": 521.46, "end": 521.7800000000001, "probability": 0.970703125 }, { "word": " your", "start": 521.7800000000001, "end": 522.0200000000001, "probability": 0.9892578125 }, { "word": " seatbelt", "start": 522.0200000000001, "end": 522.5000000000001, "probability": 0.9765625 }, { "word": " fasten", "start": 522.5000000000001, "end": 522.8400000000001, "probability": 0.1793212890625 }, { "word": " whenever", "start": 522.8400000000001, "end": 523.2, "probability": 0.9765625 }, { "word": " you", "start": 523.2, "end": 523.46, "probability": 0.99560546875 }, { "word": " are", "start": 523.46, "end": 523.6200000000001, "probability": 0.955078125 }, { "word": " seated.", "start": 523.6200000000001, "end": 523.96, "probability": 0.99951171875 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "Ladies and gentlemen, the captain has turned off the seatbelt sign. However, please keep your seatbelt fastened whenever you are seated.", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "ASR 'fasten' corrected to 'fastened' for grammatical correctness in standard announcement." }, { "start": 526.7, "end": 530.0200000000001, "text": "We expect to arrive at Tokyo International Airport.", "raw_asr_text": "We expect to arrive at Tokyo International Airport.", "source_language": "en", "source_lang": "en", "asr_packet_id": 31, "words": [], "asr_metrics": { "avg_logprob": -0.3702455759048462, "compression_ratio": 0.864406779661017, "no_speech_prob": 3.0845343640395484e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "en", "text": "We expect to arrive at Tokyo International Airport." }, { "language": "ja", "text": "We expect to arrive at Tokyo International Airport." } ], "id": 80, "review_annotation": "", "row_id": 80, "bounded_row_id": 80, "parent_row_ids": [ 80 ], "parent_row_id": 80, "packet_id": 31, "LID_evidence": [ { "region_id": 67, "probabilities": { "ja": 0.2829024404287338, "zh": 0.009741471009328961, "en": 0.602123636007309 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 67 ], "language_confidence": 0.602123636007309, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 525.9000000000001, "packet_end": 531.3, "raw_text": "We expect to arrive at Tokyo International Airport.", "selected_text": "We expect to arrive at Tokyo International Airport.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 80 ], "packet_id": 31, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "We expect to arrive at Tokyo International Airport.", "after": "We expect to arrive at Tokyo International Airport.", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard announcement phrasing matches context and ASR output perfectly." } ], "language": "en", "packet_ids": [ 31 ], "index": 80, "source_text": "We expect to arrive at Tokyo International Airport.", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Standard announcement phrasing matches context and ASR output perfectly.", "raw_word_timestamps": [ { "word": " We", "start": 525.9000000000001, "end": 527.0200000000001, "probability": 0.71337890625 }, { "word": " expect", "start": 527.0200000000001, "end": 527.2, "probability": 0.99560546875 }, { "word": " to", "start": 527.2, "end": 527.4800000000001, "probability": 0.9345703125 }, { "word": " arrive", "start": 527.4800000000001, "end": 527.7, "probability": 0.9931640625 }, { "word": " at", "start": 527.7, "end": 527.96, "probability": 0.9912109375 }, { "word": " Tokyo", "start": 527.96, "end": 528.3000000000001, "probability": 0.951171875 }, { "word": " International", "start": 528.3000000000001, "end": 529.1800000000001, "probability": 0.9443359375 }, { "word": " Airport.", "start": 529.1800000000001, "end": 530.0200000000001, "probability": 0.97509765625 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 533.4, "end": 540.1800000000001, "text": "at 5 30 p.m. March 14th", "raw_asr_text": "at 5 30 p.m. March 14th", "source_language": "en", "source_lang": "en", "asr_packet_id": 32, "words": [], "asr_metrics": { "avg_logprob": -0.5293709314786471, "compression_ratio": 0.7419354838709677, "no_speech_prob": 7.298561061075404e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "en", "text": "at 5 30 p.m. March 14th" }, { "language": "ja", "text": "5.30pmMAR 14" } ], "id": 81, "review_annotation": "", "row_id": 81, "bounded_row_id": 81, "parent_row_ids": [ 81 ], "parent_row_id": 81, "packet_id": 32, "LID_evidence": [ { "region_id": 68, "probabilities": { "ja": 0.0060543036088347435, "zh": 0.004035324789583683, "en": 0.9463245868682861 }, "status": "UNCERTAIN" }, { "region_id": 69, "probabilities": { "ja": 0.009477713843807578, "zh": 0.008119033416733146, "en": 0.6202474385499954 }, "status": "UNCERTAIN" }, { "region_id": 70, "probabilities": { "ja": 0.016617114003747702, "zh": 0.04849872766062617, "en": 0.6131685137748718 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 68, 69, 70 ], "language_confidence": 0.9463245868682861, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 532.6, "packet_end": 541.3, "raw_text": "at 5 30 p.m. March 14th", "selected_text": "at 5 30 p.m. March 14th", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 81 ], "packet_id": 32, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "at 5 30 p.m. March 14th", "after": "at 5 30 p.m. March 14th", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Clear time/date specification consistent with standard arrival announcements." } ], "language": "en", "packet_ids": [ 32 ], "index": 81, "source_text": "at 5 30 p.m. March 14th", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Clear time/date specification consistent with standard arrival announcements.", "raw_word_timestamps": [ { "word": " at", "start": 532.6, "end": 533.72, "probability": 0.71240234375 }, { "word": " 5", "start": 533.72, "end": 534.12, "probability": 0.947265625 }, { "word": " 30", "start": 534.12, "end": 534.88, "probability": 0.07867431640625 }, { "word": " p", "start": 534.88, "end": 536.96, "probability": 0.432861328125 }, { "word": ".m.", "start": 536.96, "end": 537.36, "probability": 0.9990234375 }, { "word": " March", "start": 538.58, "end": 538.86, "probability": 0.76953125 }, { "word": " 14th", "start": 538.86, "end": 540.1800000000001, "probability": 0.96826171875 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 542.9, "end": 546.84, "text": "The time difference between Songshan and Tokyo is 1 hour.", "raw_asr_text": "The time difference between Sonshan and Tokyo is 1 hour.", "source_language": "en", "source_lang": "en", "asr_packet_id": 33, "words": [], "asr_metrics": { "avg_logprob": -0.5106252801829371, "compression_ratio": 1.0975609756097562, "no_speech_prob": 1.750354201857096e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "en", "text": "The time difference between Sonshan and Tokyo is 1 hour. So local time in Tokyo is 2.42pm." }, { "language": "ja", "text": "Sonshan and Tokyo is 1hLocal time in Tokyo is 2.42pm" } ], "id": 82, "review_annotation": "", "row_id": 82, "bounded_row_id": 82, "parent_row_ids": [ 82 ], "parent_row_id": 82, "packet_id": 33, "LID_evidence": [ { "region_id": 71, "probabilities": { "ja": 0.008148714108392597, "zh": 0.003639699425548315, "en": 0.7557422757148743 }, "status": "UNCERTAIN" }, { "region_id": 72, "probabilities": { "ja": 0.04403098512973104, "zh": 0.006301587680354714, "en": 0.6740844462599073 }, "status": "STABLE" } ], "VAD_region_ids": [ 71, 72 ], "language_confidence": 0.7557422757148743, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 542.1, "packet_end": 553.5, "raw_text": "The time difference between Sonshan and Tokyo is 1 hour.", "selected_text": "The time difference between Songshan and Tokyo is 1 hour.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 82 ], "packet_id": 33, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "The time difference between Sonshan and Tokyo is 1 hour.", "after": "The time difference between Songshan and Tokyo is 1 hour.", "classification": "STRONG_RESTORATION", "reason": "ASR 'Sonshan' is a phonetic error for 'Songshan', supported by flight route context." } ], "language": "en", "packet_ids": [ 33 ], "index": 82, "independent_ASR": { "text": "The time difference between Sunshine and Tokyo is one hour, so local time in Tokyo is 2:42 p.m.", "chunks": [ { "text": "The time difference between Sunshine and Tokyo is one hour, so local time in Tokyo is 2:42 p.m.", "start": 0.0, "end": 11.4, "chunk_index": 0, "language": "English", "finish_reason": "eos", "truncated": false, "generated_tokens": 25, "max_new_tokens": 169 } ], "detected_language": "English", "segments": [ { "start": 542.1, "end": 553.5, "text": "The time difference between Sunshine and Tokyo is one hour, so local time in Tokyo is 2:42 p.m." } ], "start": 542.1, "end": 553.5, "engine": "mlx-qwen3-asr", "language": "en" }, "raw_word_timestamps": [ { "word": " The", "start": 542.1, "end": 543.12, "probability": 0.51220703125 }, { "word": " time", "start": 543.12, "end": 543.38, "probability": 0.98681640625 }, { "word": " difference", "start": 543.38, "end": 543.78, "probability": 0.9892578125 }, { "word": " between", "start": 543.78, "end": 544.22, "probability": 0.994140625 }, { "word": " Sonshan", "start": 544.22, "end": 544.8000000000001, "probability": 0.9072265625 }, { "word": " and", "start": 544.8000000000001, "end": 545.0400000000001, "probability": 0.90087890625 }, { "word": " Tokyo", "start": 545.0400000000001, "end": 546.0400000000001, "probability": 0.9775390625 }, { "word": " is", "start": 546.0400000000001, "end": 546.34, "probability": 0.98779296875 }, { "word": " 1", "start": 546.34, "end": 546.62, "probability": 0.6005859375 }, { "word": " hour.", "start": 546.62, "end": 546.84, "probability": 0.80517578125 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "The time difference between Songshan and Tokyo is 1 hour.", "review_status": "STRONG_RESTORATION", "classification": "STRONG_RESTORATION", "confidence": 0.9, "correction_reason": "ASR 'Sonshan' is a phonetic error for 'Songshan', supported by flight route context." }, { "start": 547.36, "end": 552.0400000000001, "text": "So local time in Tokyo is 2.42pm.", "raw_asr_text": "So local time in Tokyo is 2.42pm.", "source_language": "en", "source_lang": "en", "asr_packet_id": 33, "words": [ { "word": " So", "start": 547.36, "end": 547.84, "probability": 0.6103515625 }, { "word": " local", "start": 547.84, "end": 548.1, "probability": 0.81494140625 }, { "word": " time", "start": 548.1, "end": 548.4200000000001, "probability": 0.99658203125 }, { "word": " in", "start": 548.4200000000001, "end": 548.7, "probability": 0.9892578125 }, { "word": " Tokyo", "start": 548.7, "end": 549.0600000000001, "probability": 0.9892578125 }, { "word": " is", "start": 549.0600000000001, "end": 549.46, "probability": 0.9951171875 }, { "word": " 2", "start": 549.46, "end": 550.4, "probability": 0.95556640625 }, { "word": ".42pm.", "start": 550.4, "end": 552.0400000000001, "probability": 0.494140625 } ], "asr_metrics": { "avg_logprob": -0.5106252801829371, "compression_ratio": 1.0975609756097562, "no_speech_prob": 1.750354201857096e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "en", "text": "The time difference between Sonshan and Tokyo is 1 hour. So local time in Tokyo is 2.42pm." }, { "language": "ja", "text": "Sonshan and Tokyo is 1hLocal time in Tokyo is 2.42pm" } ], "id": 83, "review_annotation": "", "row_id": 83, "bounded_row_id": 83, "parent_row_ids": [ 83 ], "parent_row_id": 83, "packet_id": 33, "LID_evidence": [ { "region_id": 73, "probabilities": { "ja": 0.23166349926032126, "zh": 0.006300873195868917, "en": 0.6035042218863964 }, "status": "STABLE" }, { "region_id": 74, "probabilities": { "ja": 0.158083476126194, "zh": 0.005642178099757682, "en": 0.7352907458941141 }, "status": "STABLE" } ], "VAD_region_ids": [ 73, 74 ], "language_confidence": 0.7352907458941141, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 542.1, "packet_end": 553.5, "raw_text": "So local time in Tokyo is 2.42pm.", "selected_text": "So local time in Tokyo is 2.42pm.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 83 ], "packet_id": 33, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "So local time in Tokyo is 2.42pm.", "after": "So local time in Tokyo is 2.42pm.", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Consistent with preceding time context and standard announcement structure." } ], "language": "en", "packet_ids": [ 33 ], "index": 83, "source_text": "So local time in Tokyo is 2.42pm.", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Consistent with preceding time context and standard announcement structure." }, { "start": 559.7, "end": 560.5, "text": "[聴不清／待確認]", "raw_asr_text": "", "source_language": "ja", "source_lang": "ja", "vad_region_id": 75, "review_status": "UNRECOVERABLE_MARKED", "asr_packet_id": 34, "correction_reason": "ASR output is incoherent; no responsible lexical recovery exists.", "confidence": 0.1, "id": 84, "review_annotation": "待確認", "row_id": 84, "bounded_row_id": 84, "parent_row_ids": [ 84 ], "parent_row_id": 84, "packet_id": 34, "LID_evidence": [ { "region_id": 75, "probabilities": { "ja": 0, "zh": 0, "en": 0 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 75 ], "language_confidence": 0, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 558.9000000000001, "packet_end": 575.6999999999999, "raw_text": "", "selected_text": "[聴不清／待確認]", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 84 ], "packet_id": 34, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "[聴不清／待確認]", "after": "[聴不清／待確認]", "classification": "UNRECOVERABLE_MARKED", "reason": "ASR output is incoherent; no responsible lexical recovery exists." } ], "language": "ja", "packet_ids": [ 34 ], "index": 84, "independent_ASR": { "engine": "kotoba-whisper", "model_path": "/Users/sagawa/AI_WORK/cache/huggingface/hub/models--kaiinui--kotoba-whisper-v2.0-mlx/snapshots/1a5ac6e52394898ef8996212e849d964e0dfb9ae", "language": "ja", "audio_interval": [ 558.9000000000001, 575.6999999999999 ], "target_interval": [ 559.7, 574.9 ], "text": "メイレーリー、レイル、レイルスクリーム、エアソース、エリース、エリース、!エリースクリア、クレース、!!!オリア、!!や、!!", "words": [ { "word": "メ", "start": 559.6000000000001, "end": 560.0400000000001, "probability": 0.481201171875 }, { "word": "イ", "start": 560.0400000000001, "end": 560.0600000000001, "probability": 0.65283203125 }, { "word": "レ", "start": 560.0600000000001, "end": 560.3400000000001, "probability": 0.525390625 }, { "word": "ー", "start": 560.3400000000001, "end": 560.3400000000001, "probability": 0.744140625 }, { "word": "リ", "start": 560.3400000000001, "end": 562.4800000000001, "probability": 0.51611328125 }, { "word": "ー、", "start": 562.4800000000001, "end": 562.4800000000001, "probability": 0.98583984375 }, { "word": "レ", "start": 562.9000000000001, "end": 563.3000000000001, "probability": 0.6171875 }, { "word": "イ", "start": 563.3000000000001, "end": 563.3400000000001, "probability": 0.84130859375 }, { "word": "ル、", "start": 563.3400000000001, "end": 563.4800000000001, "probability": 0.9033203125 }, { "word": "レ", "start": 563.5200000000001, "end": 563.94, "probability": 0.160888671875 }, { "word": "イ", "start": 563.94, "end": 563.94, "probability": 0.7587890625 }, { "word": "ル", "start": 563.94, "end": 563.94, "probability": 0.75244140625 }, { "word": "ス", "start": 563.94, "end": 564.0600000000001, "probability": 0.984375 }, { "word": "ク", "start": 564.0600000000001, "end": 564.2400000000001, "probability": 0.9375 }, { "word": "リ", "start": 564.2400000000001, "end": 564.32, "probability": 0.81689453125 }, { "word": "ーム、", "start": 564.32, "end": 564.46, "probability": 0.99755859375 }, { "word": "エ", "start": 564.6800000000001, "end": 565.2, "probability": 0.62841796875 }, { "word": "ア", "start": 565.2, "end": 565.2, "probability": 0.258544921875 }, { "word": "ソ", "start": 565.2, "end": 565.3400000000001, "probability": 0.376708984375 }, { "word": "ー", "start": 565.3400000000001, "end": 565.4000000000001, "probability": 0.619140625 }, { "word": "ス、", "start": 565.4000000000001, "end": 565.5000000000001, "probability": 0.99853515625 }, { "word": "エ", "start": 566.0200000000001, "end": 566.32, "probability": 0.77099609375 }, { "word": "リ", "start": 566.32, "end": 566.32, "probability": 0.1510009765625 }, { "word": "ー", "start": 566.32, "end": 566.32, "probability": 0.62060546875 }, { "word": "ス、", "start": 566.32, "end": 566.32, "probability": 0.68896484375 }, { "word": "エ", "start": 566.32, "end": 566.32, "probability": 0.56396484375 }, { "word": "リ", "start": 566.32, "end": 566.32, "probability": 0.1993408203125 }, { "word": "ー", "start": 566.32, "end": 566.32, "probability": 0.85595703125 }, { "word": "ス、!", "start": 566.32, "end": 566.4000000000001, "probability": 0.5888671875 }, { "word": "エ", "start": 567.46, "end": 567.46, "probability": 0.21240234375 }, { "word": "リ", "start": 567.46, "end": 567.46, "probability": 0.2054443359375 }, { "word": "ー", "start": 567.46, "end": 567.46, "probability": 0.609375 }, { "word": "ス", "start": 567.46, "end": 567.8400000000001, "probability": 0.89208984375 }, { "word": "ク", "start": 567.8400000000001, "end": 567.8400000000001, "probability": 0.2139892578125 }, { "word": "リ", "start": 567.8400000000001, "end": 568.1800000000001, "probability": 0.6083984375 }, { "word": "ア、", "start": 568.1800000000001, "end": 568.8800000000001, "probability": 0.310791015625 }, { "word": "ク", "start": 568.8800000000001, "end": 569.1400000000001, "probability": 0.1673583984375 }, { "word": "レ", "start": 569.1400000000001, "end": 569.1400000000001, "probability": 0.45166015625 }, { "word": "ー", "start": 569.1400000000001, "end": 569.1400000000001, "probability": 0.73876953125 }, { "word": "ス、!!!", "start": 569.1400000000001, "end": 569.2800000000001, "probability": 0.280517578125 }, { "word": "オ", "start": 569.5000000000001, "end": 569.5200000000001, "probability": 0.059783935546875 }, { "word": "リ", "start": 569.5200000000001, "end": 570.0400000000001, "probability": 0.62109375 }, { "word": "ア、!!", "start": 570.0400000000001, "end": 570.1400000000001, "probability": 0.312255859375 }, { "word": "や、!!", "start": 571.6600000000001, "end": 571.6600000000001, "probability": 0.110107421875 } ], "raw_result": { "text": "メイレーリー、レイル、レイルスクリーム、エアソース、エリース、エリース、!エリースクリア、クレース、!!!オリア、!!や、!", "segments": [ { "id": 0, "seek": 0, "start": 0.6999999999999986, "end": 6.6, "text": "メイレーリー、レイル、レイルスクリーム、エアソース、", "tokens": [ 50365, 21647, 8040, 16680, 3384, 12376, 3384, 1231, 16680, 8040, 9405, 1231, 16680, 8040, 9405, 9550, 10825, 12376, 44265, 1231, 26215, 12817, 42668, 3384, 9550, 1231, 50697 ], "temperature": 0.0, "avg_logprob": NaN, "compression_ratio": 1.9438202247191012, "no_speech_prob": 0.06888698786497116, "words": [ { "word": "メ", "start": 0.6999999999999986, "end": 1.14, "probability": 0.481201171875 }, { "word": "イ", "start": 1.14, "end": 1.16, "probability": 0.65283203125 }, { "word": "レ", "start": 1.16, "end": 1.44, "probability": 0.525390625 }, { "word": "ー", "start": 1.44, "end": 1.44, "probability": 0.744140625 }, { "word": "リ", "start": 1.44, "end": 3.58, "probability": 0.51611328125 }, { "word": "ー、", "start": 3.58, "end": 3.58, "probability": 0.98583984375 }, { "word": "レ", "start": 4.0, "end": 4.4, "probability": 0.6171875 }, { "word": "イ", "start": 4.4, "end": 4.44, "probability": 0.84130859375 }, { "word": "ル、", "start": 4.44, "end": 4.58, "probability": 0.9033203125 }, { "word": "レ", "start": 4.62, "end": 5.04, "probability": 0.160888671875 }, { "word": "イ", "start": 5.04, "end": 5.04, "probability": 0.7587890625 }, { "word": "ル", "start": 5.04, "end": 5.04, "probability": 0.75244140625 }, { "word": "ス", "start": 5.04, "end": 5.16, "probability": 0.984375 }, { "word": "ク", "start": 5.16, "end": 5.34, "probability": 0.9375 }, { "word": "リ", "start": 5.34, "end": 5.42, "probability": 0.81689453125 }, { "word": "ーム、", "start": 5.42, "end": 5.56, "probability": 0.99755859375 }, { "word": "エ", "start": 5.78, "end": 6.3, "probability": 0.62841796875 }, { "word": "ア", "start": 6.3, "end": 6.3, "probability": 0.258544921875 }, { "word": "ソ", "start": 6.3, "end": 6.44, "probability": 0.376708984375 }, { "word": "ー", "start": 6.44, "end": 6.5, "probability": 0.619140625 }, { "word": "ス、", "start": 6.5, "end": 6.6, "probability": 0.99853515625 } ] }, { "id": 1, "seek": 0, "start": 7.12, "end": 7.5, "text": "エリース、エリース、", "tokens": [ 50697, 26215, 12376, 3384, 9550, 1231, 26215, 12376, 3384, 9550, 1231, 50696 ], "temperature": 0.0, "avg_logprob": NaN, "compression_ratio": 1.9438202247191012, "no_speech_prob": 0.06888698786497116, "words": [ { "word": "エ", "start": 7.12, "end": 7.42, "probability": 0.77099609375 }, { "word": "リ", "start": 7.42, "end": 7.42, "probability": 0.1510009765625 }, { "word": "ー", "start": 7.42, "end": 7.42, "probability": 0.62060546875 }, { "word": "ス、", "start": 7.42, "end": 7.42, "probability": 0.68896484375 }, { "word": "エ", "start": 7.42, "end": 7.42, "probability": 0.56396484375 }, { "word": "リ", "start": 7.42, "end": 7.42, "probability": 0.1993408203125 }, { "word": "ー", "start": 7.42, "end": 7.42, "probability": 0.85595703125 }, { "word": "ス、!", "start": 7.42, "end": 7.5, "probability": 0.5888671875 } ] }, { "id": 2, "seek": 0, "start": 8.56, "end": 10.38, "text": "!エリースクリア、クレース、", "tokens": [ 50696, 0, 26215, 12376, 3384, 9550, 10825, 12376, 12817, 1231, 10825, 16680, 3384, 9550, 1231, 50838 ], "temperature": 0.0, "avg_logprob": NaN, "compression_ratio": 1.9438202247191012, "no_speech_prob": 0.06888698786497116, "words": [ { "word": "エ", "start": 8.56, "end": 8.56, "probability": 0.21240234375 }, { "word": "リ", "start": 8.56, "end": 8.56, "probability": 0.2054443359375 }, { "word": "ー", "start": 8.56, "end": 8.56, "probability": 0.609375 }, { "word": "ス", "start": 8.56, "end": 8.94, "probability": 0.89208984375 }, { "word": "ク", "start": 8.94, "end": 8.94, "probability": 0.2139892578125 }, { "word": "リ", "start": 8.94, "end": 9.28, "probability": 0.6083984375 }, { "word": "ア、", "start": 9.28, "end": 9.98, "probability": 0.310791015625 }, { "word": "ク", "start": 9.98, "end": 10.24, "probability": 0.1673583984375 }, { "word": "レ", "start": 10.24, "end": 10.24, "probability": 0.45166015625 }, { "word": "ー", "start": 10.24, "end": 10.24, "probability": 0.73876953125 }, { "word": "ス、!!!", "start": 10.24, "end": 10.38, "probability": 0.280517578125 } ] }, { "id": 3, "seek": 0, "start": 10.6, "end": 10.62, "text": "!", "tokens": [ 50838, 0, 50792 ], "temperature": 0.0, "avg_logprob": NaN, "compression_ratio": 1.9438202247191012, "no_speech_prob": 0.06888698786497116, "words": [ { "word": "オ", "start": 10.6, "end": 10.62, "probability": 0.059783935546875 } ] }, { "id": 4, "seek": 0, "start": 10.62, "end": 11.14, "text": "!", "tokens": [ 50792, 0, 50745 ], "temperature": 0.0, "avg_logprob": NaN, "compression_ratio": 1.9438202247191012, "no_speech_prob": 0.06888698786497116, "words": [ { "word": "リ", "start": 10.62, "end": 11.14, "probability": 0.62109375 } ] }, { "id": 5, "seek": 0, "start": 11.14, "end": 12.76, "text": "!オリア、", "tokens": [ 50745, 0, 18743, 12376, 12817, 1231, 51017 ], "temperature": 0.0, "avg_logprob": NaN, "compression_ratio": 1.9438202247191012, "no_speech_prob": 0.06888698786497116, "words": [ { "word": "ア、!!", "start": 11.14, "end": 11.24, "probability": 0.312255859375 }, { "word": "や、!!", "start": 12.76, "end": 12.76, "probability": 0.110107421875 } ] }, { "id": 6, "seek": 0, "start": 13.040000000000001, "end": 15.58, "text": "!", "tokens": [ 51017, 0, 51144 ], "temperature": 0.0, "avg_logprob": NaN, "compression_ratio": 1.9438202247191012, "no_speech_prob": 0.06888698786497116, "words": [] }, { "id": 7, "seek": 0, "start": 15.58, "end": 14.84, "text": "!や、", "tokens": [ 51144, 0, 7355, 1231, 51107 ], "temperature": 0.0, "avg_logprob": NaN, "compression_ratio": 1.9438202247191012, "no_speech_prob": 0.06888698786497116, "words": [] }, { "id": 8, "seek": 0, "start": 14.84, "end": 15.64, "text": "!", "tokens": [ 51107, 0, 51147 ], "temperature": 0.0, "avg_logprob": NaN, "compression_ratio": 1.9438202247191012, "no_speech_prob": 0.06888698786497116, "words": [] }, { "id": 9, "seek": 0, "start": 15.64, "end": 15.64, "text": "", "tokens": [], "temperature": 0.0, "avg_logprob": NaN, "compression_ratio": 1.9438202247191012, "no_speech_prob": 0.06888698786497116, "words": [] } ], "language": "ja", "duration": 16.8 }, "status": "COMPLETE", "rejected_alternative": { "text": "May be, may be reminded that liquids, gels, creams, and aerosols exceeding 100 milliliters capacity purchased before departure are not allowed on board connecting international flight or return flight.", "chunks": [ { "text": "May be, may be reminded that liquids, gels, creams, and aerosols exceeding 100 milliliters capacity purchased before departure are not allowed on board connecting international flight or return flight.", "start": 0.0, "end": 16.8, "chunk_index": 0, "language": "Japanese", "finish_reason": "eos", "truncated": false, "generated_tokens": 41, "max_new_tokens": 234 } ], "detected_language": "Japanese", "segments": [ { "start": 558.9000000000001, "end": 575.7, "text": "May be, may be reminded that liquids, gels, creams, and aerosols exceeding 100 milliliters capacity purchased before departure are not allowed on board connecting international flight or return flight." } ], "start": 558.9000000000001, "end": 575.6999999999999, "engine": "mlx-qwen3-asr", "language": "ja" } }, "source_text": "[聴不清／待確認]", "classification": "UNRECOVERABLE_MARKED" }, { "start": 561.1600000000001, "end": 574.46, "text": "May we be reminded that liquids, gels, creams, and aerosols exceeding 100 ml capacity purchased before departure are not allowed on board yet connecting into national flight or return flight.", "raw_asr_text": "May we be reminded that liquids, gels, creams, and aerosols exceeding 100 ml capacity purchased before departure are not allowed on board yet connecting into national flight or return flight.", "source_language": "en", "source_lang": "en", "asr_packet_id": 34, "words": [ { "word": " May", "start": 561.1600000000001, "end": 561.9200000000001, "probability": 0.708984375 }, { "word": " we", "start": 561.9200000000001, "end": 562.0600000000001, "probability": 0.96240234375 }, { "word": " be", "start": 562.0600000000001, "end": 562.1200000000001, "probability": 0.321533203125 }, { "word": " reminded", "start": 562.1200000000001, "end": 562.5400000000001, "probability": 0.99365234375 }, { "word": " that", "start": 562.5400000000001, "end": 562.8600000000001, "probability": 0.98388671875 }, { "word": " liquids,", "start": 562.8600000000001, "end": 563.3600000000001, "probability": 0.8916015625 }, { "word": " gels,", "start": 563.6000000000001, "end": 563.8600000000001, "probability": 0.93505859375 }, { "word": " creams,", "start": 564.1200000000001, "end": 564.3800000000001, "probability": 0.98486328125 }, { "word": " and", "start": 564.58, "end": 564.7200000000001, "probability": 0.96435546875 }, { "word": " aerosols", "start": 564.7200000000001, "end": 565.4800000000001, "probability": 0.953125 }, { "word": " exceeding", "start": 565.4800000000001, "end": 566.44, "probability": 0.97265625 }, { "word": " 100", "start": 566.44, "end": 567.0000000000001, "probability": 0.97216796875 }, { "word": " ml", "start": 567.0000000000001, "end": 567.4200000000001, "probability": 0.2113037109375 }, { "word": " capacity", "start": 567.4200000000001, "end": 568.2200000000001, "probability": 0.4228515625 }, { "word": " purchased", "start": 568.2200000000001, "end": 569.1800000000001, "probability": 0.93798828125 }, { "word": " before", "start": 569.1800000000001, "end": 569.6000000000001, "probability": 0.998046875 }, { "word": " departure", "start": 569.6000000000001, "end": 570.1400000000001, "probability": 0.99755859375 }, { "word": " are", "start": 570.1400000000001, "end": 570.6000000000001, "probability": 0.9853515625 }, { "word": " not", "start": 570.6000000000001, "end": 570.8400000000001, "probability": 0.9931640625 }, { "word": " allowed", "start": 570.8400000000001, "end": 571.2, "probability": 0.93994140625 }, { "word": " on", "start": 571.2, "end": 571.3800000000001, "probability": 0.56689453125 }, { "word": " board", "start": 571.3800000000001, "end": 571.58, "probability": 0.734375 }, { "word": " yet", "start": 571.58, "end": 571.7200000000001, "probability": 0.90234375 }, { "word": " connecting", "start": 571.7200000000001, "end": 572.2, "probability": 0.94091796875 }, { "word": " into", "start": 572.2, "end": 572.6400000000001, "probability": 0.5791015625 }, { "word": " national", "start": 572.6400000000001, "end": 573.0400000000001, "probability": 0.869140625 }, { "word": " flight", "start": 573.0400000000001, "end": 573.5000000000001, "probability": 0.97216796875 }, { "word": " or", "start": 573.5000000000001, "end": 573.7800000000001, "probability": 0.9912109375 }, { "word": " return", "start": 573.7800000000001, "end": 574.0600000000001, "probability": 0.96435546875 }, { "word": " flight.", "start": 574.0600000000001, "end": 574.46, "probability": 0.98876953125 } ], "asr_metrics": { "avg_logprob": -0.28736412525177, "compression_ratio": 1.3546099290780143, "no_speech_prob": 3.268693649083332e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "en", "text": "May we be reminded that liquids, gels, creams, and aerosols exceeding 100 ml capacity purchased before departure are not allowed on board yet connecting into national flight or return flight." }, { "language": "ja", "text": "また、水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道の水道" } ], "id": 85, "review_annotation": "待確認", "row_id": 85, "bounded_row_id": 85, "parent_row_ids": [ 85 ], "parent_row_id": 85, "packet_id": 34, "LID_evidence": [ { "region_id": 76, "probabilities": { "ja": 0.04066248511662707, "zh": 0.0050426610820674496, "en": 0.8769301995635033 }, "status": "STABLE" } ], "VAD_region_ids": [ 76 ], "language_confidence": 0.8769301995635033, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 558.9000000000001, "packet_end": 575.6999999999999, "raw_text": "May we be reminded that liquids, gels, creams, and aerosols exceeding 100 ml capacity purchased before departure are not allowed on board yet connecting into national flight or return flight.", "selected_text": "May we be reminded that liquids, gels, creams, and aerosols exceeding 100 ml capacity purchased before departure are not allowed on board yet connecting into national flight or return flight.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 85 ], "packet_id": 34, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "May we be reminded that liquids, gels, creams, and aerosols exceeding 100 ml capacity purchased before departure are not allowed on board yet connecting into national flight or return flight.", "after": "May we be reminded that liquids, gels, creams, and aerosols exceeding 100 ml capacity purchased before departure are not allowed on board yet connecting into national flight or return flight.", "classification": "BEST_SUPPORTED_WITH_REVIEW", "reason": "Independent ASR confirms 'international' over 'national' and 'connecting' phrasing; 'yet' likely ASR error for 'when' or 'while'." } ], "language": "en", "packet_ids": [ 34 ], "index": 85, "independent_ASR": { "text": "May we may be reminded that liquids, gels, creams, and aerosols exceeding 100 milliliters capacity purchased before departure are not allowed on board, connecting international flight or return flight.", "chunks": [ { "text": "May we may be reminded that liquids, gels, creams, and aerosols exceeding 100 milliliters capacity purchased before departure are not allowed on board, connecting international flight or return flight.", "start": 0.0, "end": 16.8, "chunk_index": 0, "language": "English", "finish_reason": "eos", "truncated": false, "generated_tokens": 41, "max_new_tokens": 234 } ], "detected_language": "English", "segments": [ { "start": 558.9000000000001, "end": 575.7, "text": "May we may be reminded that liquids, gels, creams, and aerosols exceeding 100 milliliters capacity purchased before departure are not allowed on board, connecting international flight or return flight." } ], "start": 558.9000000000001, "end": 575.6999999999999, "engine": "mlx-qwen3-asr", "language": "en" }, "source_text": "May we be reminded that liquids, gels, creams, and aerosols exceeding 100 ml capacity purchased before departure are not allowed on board yet connecting into national flight or return flight.", "review_status": "BEST_SUPPORTED_WITH_REVIEW", "classification": "BEST_SUPPORTED_WITH_REVIEW", "confidence": 0.85, "correction_reason": "Independent ASR confirms 'international' over 'national' and 'connecting' phrasing; 'yet' likely ASR error for 'when' or 'while'." }, { "start": 581.1800000000001, "end": 590.3000000000001, "text": "Passengers, if you have any symptoms such as fever or stomach ache, please inform a cabin attendant. Thank you.", "raw_asr_text": "Passengers, if you have any symptoms such as fever or stomach ache, please inform a cabin attendant. Thank you.", "source_language": "en", "source_lang": "en", "asr_packet_id": 35, "words": [ { "word": " Passengers,", "start": 581.1800000000001, "end": 581.9000000000001, "probability": 0.75 }, { "word": " if", "start": 582.8000000000001, "end": 582.8800000000001, "probability": 0.962890625 }, { "word": " you", "start": 582.8800000000001, "end": 583.0000000000001, "probability": 0.99853515625 }, { "word": " have", "start": 583.0000000000001, "end": 583.2800000000001, "probability": 0.9970703125 }, { "word": " any", "start": 583.2800000000001, "end": 583.98, "probability": 0.97314453125 }, { "word": " symptoms", "start": 583.98, "end": 584.58, "probability": 0.99560546875 }, { "word": " such", "start": 584.58, "end": 585.2600000000001, "probability": 0.76416015625 }, { "word": " as", "start": 585.2600000000001, "end": 585.48, "probability": 0.9990234375 }, { "word": " fever", "start": 585.48, "end": 585.86, "probability": 0.98583984375 }, { "word": " or", "start": 585.86, "end": 586.1600000000001, "probability": 0.990234375 }, { "word": " stomach", "start": 586.1600000000001, "end": 586.5200000000001, "probability": 0.99267578125 }, { "word": " ache,", "start": 586.5200000000001, "end": 587.0400000000001, "probability": 0.75341796875 }, { "word": " please", "start": 587.2800000000001, "end": 587.8000000000001, "probability": 0.994140625 }, { "word": " inform", "start": 587.8000000000001, "end": 588.2400000000001, "probability": 0.974609375 }, { "word": " a", "start": 588.2400000000001, "end": 588.44, "probability": 0.429443359375 }, { "word": " cabin", "start": 588.44, "end": 588.7600000000001, "probability": 0.99267578125 }, { "word": " attendant.", "start": 588.7600000000001, "end": 589.34, "probability": 0.99951171875 }, { "word": " Thank", "start": 589.6800000000001, "end": 590.0400000000001, "probability": 0.99462890625 }, { "word": " you.", "start": 590.0400000000001, "end": 590.3000000000001, "probability": 0.9736328125 } ], "asr_metrics": { "avg_logprob": -0.2846373098867911, "compression_ratio": 1.168421052631579, "no_speech_prob": 4.4679218658139064e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "en", "text": "Passengers, if you have any symptoms such as fever or stomach ache, please inform a cabin attendant. Thank you." } ], "id": 86, "review_annotation": "", "row_id": 86, "bounded_row_id": 86, "parent_row_ids": [ 86 ], "parent_row_id": 86, "packet_id": 35, "LID_evidence": [ { "region_id": 77, "probabilities": { "ja": 0.03587866201996803, "zh": 0.004700699433063467, "en": 0.9166288177172343 }, "status": "STABLE" }, { "region_id": 78, "probabilities": { "ja": 0.031159180767225243, "zh": 0.0038044264880367196, "en": 0.8937849677526034 }, "status": "STABLE" }, { "region_id": 79, "probabilities": { "ja": 0.040188660246453116, "zh": 0.004927500234251576, "en": 0.8457913398742676 }, "status": "STABLE" } ], "VAD_region_ids": [ 77, 78, 79 ], "language_confidence": 0.9166288177172343, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 580.3000000000001, "packet_end": 591.5, "raw_text": "Passengers, if you have any symptoms such as fever or stomach ache, please inform a cabin attendant. Thank you.", "selected_text": "Passengers, if you have any symptoms such as fever or stomach ache, please inform a cabin attendant. Thank you.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 86 ], "packet_id": 35, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "Passengers, if you have any symptoms such as fever or stomach ache, please inform a cabin attendant. Thank you.", "after": "Passengers, if you have any symptoms such as fever or stomach ache, please inform a cabin attendant. Thank you.", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard health advisory phrasing matches ASR and context perfectly." } ], "language": "en", "packet_ids": [ 35 ], "index": 86, "source_text": "Passengers, if you have any symptoms such as fever or stomach ache, please inform a cabin attendant. Thank you.", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Standard health advisory phrasing matches ASR and context perfectly." }, { "start": 595.1, "end": 601.84, "text": "各位贵宾,本班機预计在当地时间下午5点30分抵达东京羽田国际机场", "raw_asr_text": "各位贵宾,本班季预计在当地时间下午5点30分抵达东京与天国际机场", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 36, "words": [], "asr_metrics": { "avg_logprob": -0.21198173308036697, "compression_ratio": 1.1597222222222223, "no_speech_prob": 2.3566499446348033e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "各位贵宾,本班季预计在当地时间下午5点30分抵达东京与天国际机场现在当地的时间为下午2点42分台北和东京之间的时差为1小时" } ], "id": 87, "review_annotation": "", "row_id": 87, "bounded_row_id": 87, "parent_row_ids": [ 87 ], "parent_row_id": 87, "packet_id": 36, "LID_evidence": [ { "region_id": 80, "probabilities": { "ja": 0.017274811508832498, "zh": 0.8792548795541127, "en": 0.06446547429077328 }, "status": "STABLE" } ], "VAD_region_ids": [ 80 ], "language_confidence": 0.8792548795541127, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 594.3000000000001, "packet_end": 612.1999999999999, "raw_text": "各位贵宾,本班季预计在当地时间下午5点30分抵达东京与天国际机场", "selected_text": "各位贵宾,本班機预计在当地时间下午5点30分抵达东京羽田国际机场", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 87 ], "packet_id": 36, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "各位贵宾,本班季预计在当地时间下午5点30分抵达东京与天国际机场", "after": "各位贵宾,本班機预计在当地时间下午5点30分抵达东京羽田国际机场", "classification": "STRONG_RESTORATION", "reason": "Independent ASR confirms '機' (flight) and '羽田' (Haneda); '季' and '与天' are clear ASR errors for flight number and airport name." } ], "language": "zh", "packet_ids": [ 36 ], "index": 87, "independent_ASR": { "text": "各位贵宾，本班机预计在当地时间下午五点三十分抵达东京羽田国际机场。现在当地的时间为下午两点四十二分，台北和和东京之间的时差为一小时。", "chunks": [ { "text": "各位贵宾，本班机预计在当地时间下午五点三十分抵达东京羽田国际机场。现在当地的时间为下午两点四十二分，台北和和东京之间的时差为一小时。", "start": 0.0, "end": 17.9, "chunk_index": 0, "language": "Chinese", "finish_reason": "eos", "truncated": false, "generated_tokens": 41, "max_new_tokens": 247 } ], "detected_language": "Chinese", "segments": [ { "start": 594.3000000000001, "end": 612.2, "text": "各位贵宾，本班机预计在当地时间下午五点三十分抵达东京羽田国际机场。现在当地的时间为下午两点四十二分，台北和和东京之间的时差为一小时。" } ], "start": 594.3000000000001, "end": 612.1999999999999, "engine": "mlx-qwen3-asr", "language": "zh" }, "raw_word_timestamps": [ { "word": "各位", "start": 594.96, "end": 595.4000000000001, "probability": 0.94873046875 }, { "word": "贵", "start": 595.4000000000001, "end": 595.6, "probability": 0.78857421875 }, { "word": "宾,", "start": 595.6, "end": 595.82, "probability": 0.9716796875 }, { "word": "本", "start": 596.0400000000001, "end": 596.44, "probability": 0.9814453125 }, { "word": "班", "start": 596.44, "end": 596.6, "probability": 0.61181640625 }, { "word": "季", "start": 596.6, "end": 596.82, "probability": 0.69921875 }, { "word": "预", "start": 596.82, "end": 597.0000000000001, "probability": 0.974609375 }, { "word": "计", "start": 597.0000000000001, "end": 597.1600000000001, "probability": 0.97314453125 }, { "word": "在", "start": 597.1600000000001, "end": 597.3800000000001, "probability": 0.98388671875 }, { "word": "当", "start": 597.3800000000001, "end": 597.84, "probability": 0.9814453125 }, { "word": "地", "start": 597.84, "end": 598.0400000000001, "probability": 0.9990234375 }, { "word": "时间", "start": 598.0400000000001, "end": 598.4000000000001, "probability": 0.96728515625 }, { "word": "下", "start": 598.4000000000001, "end": 598.82, "probability": 0.9482421875 }, { "word": "午", "start": 598.82, "end": 599.0400000000001, "probability": 0.99853515625 }, { "word": "5", "start": 599.0400000000001, "end": 599.2600000000001, "probability": 0.94873046875 }, { "word": "点", "start": 599.2600000000001, "end": 599.4200000000001, "probability": 0.8037109375 }, { "word": "30", "start": 599.4200000000001, "end": 599.7400000000001, "probability": 0.99609375 }, { "word": "分", "start": 599.7400000000001, "end": 600.0000000000001, "probability": 0.998046875 }, { "word": "抵", "start": 600.0000000000001, "end": 600.1800000000001, "probability": 0.90771484375 }, { "word": "达", "start": 600.1800000000001, "end": 600.4000000000001, "probability": 0.9990234375 }, { "word": "东", "start": 600.4000000000001, "end": 600.58, "probability": 0.9931640625 }, { "word": "京", "start": 600.58, "end": 600.8000000000001, "probability": 0.97216796875 }, { "word": "与", "start": 600.8000000000001, "end": 600.98, "probability": 0.58984375 }, { "word": "天", "start": 600.98, "end": 601.08, "probability": 0.5732421875 }, { "word": "国", "start": 601.08, "end": 601.2400000000001, "probability": 0.99853515625 }, { "word": "际", "start": 601.2400000000001, "end": 601.4000000000001, "probability": 0.904296875 }, { "word": "机", "start": 601.4000000000001, "end": 601.5200000000001, "probability": 0.99169921875 }, { "word": "场", "start": 601.5200000000001, "end": 601.84, "probability": 0.9990234375 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "各位贵宾,本班機预计在当地时间下午5点30分抵达东京羽田国际机场", "review_status": "STRONG_RESTORATION", "classification": "STRONG_RESTORATION", "confidence": 0.9, "correction_reason": "Independent ASR confirms '機' (flight) and '羽田' (Haneda); '季' and '与天' are clear ASR errors for flight number and airport name." }, { "start": 602.94, "end": 606.6400000000001, "text": "现在当地的时间为下午2点42分", "raw_asr_text": "现在当地的时间为下午2点42分", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 36, "words": [ { "word": "现在", "start": 602.94, "end": 603.3800000000001, "probability": 0.75732421875 }, { "word": "当", "start": 603.3800000000001, "end": 603.6200000000001, "probability": 0.9951171875 }, { "word": "地", "start": 603.6200000000001, "end": 603.7800000000001, "probability": 0.99951171875 }, { "word": "的", "start": 603.7800000000001, "end": 603.9000000000001, "probability": 0.97119140625 }, { "word": "时间", "start": 603.9000000000001, "end": 604.2600000000001, "probability": 0.99853515625 }, { "word": "为", "start": 604.2600000000001, "end": 604.6200000000001, "probability": 0.9794921875 }, { "word": "下", "start": 604.6200000000001, "end": 605.0600000000001, "probability": 0.9853515625 }, { "word": "午", "start": 605.0600000000001, "end": 605.3000000000001, "probability": 1.0 }, { "word": "2", "start": 605.3000000000001, "end": 605.5600000000001, "probability": 0.9736328125 }, { "word": "点", "start": 605.5600000000001, "end": 605.8000000000001, "probability": 0.99951171875 }, { "word": "42", "start": 605.8000000000001, "end": 606.1600000000001, "probability": 0.99658203125 }, { "word": "分", "start": 606.1600000000001, "end": 606.6400000000001, "probability": 0.99951171875 } ], "asr_metrics": { "avg_logprob": -0.21198173308036697, "compression_ratio": 1.1597222222222223, "no_speech_prob": 2.3566499446348033e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "各位贵宾,本班季预计在当地时间下午5点30分抵达东京与天国际机场现在当地的时间为下午2点42分台北和东京之间的时差为1小时" } ], "id": 88, "review_annotation": "", "row_id": 88, "bounded_row_id": 88, "parent_row_ids": [ 88 ], "parent_row_id": 88, "packet_id": 36, "LID_evidence": [ { "region_id": 81, "probabilities": { "ja": 0.005624283427922902, "zh": 0.8843749165534973, "en": 0.071543275505643 }, "status": "STABLE" } ], "VAD_region_ids": [ 81 ], "language_confidence": 0.8843749165534973, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 594.3000000000001, "packet_end": 612.1999999999999, "raw_text": "现在当地的时间为下午2点42分", "selected_text": "现在当地的时间为下午2点42分", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 88 ], "packet_id": 36, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "现在当地的时间为下午2点42分", "after": "现在当地的时间为下午2点42分", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Clear time statement consistent with English transcript and standard announcements." } ], "language": "zh", "packet_ids": [ 36 ], "index": 88, "source_text": "现在当地的时间为下午2点42分", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Clear time statement consistent with English transcript and standard announcements." }, { "start": 607.72, "end": 610.9200000000001, "text": "台北和东京之间的时差为1小时", "raw_asr_text": "台北和东京之间的时差为1小时", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 36, "words": [ { "word": "台", "start": 607.72, "end": 608.1600000000001, "probability": 0.98681640625 }, { "word": "北", "start": 608.1600000000001, "end": 608.34, "probability": 1.0 }, { "word": "和", "start": 608.34, "end": 608.5400000000001, "probability": 0.96728515625 }, { "word": "东", "start": 608.5400000000001, "end": 609.1400000000001, "probability": 0.99755859375 }, { "word": "京", "start": 609.1400000000001, "end": 609.3000000000001, "probability": 0.99951171875 }, { "word": "之", "start": 609.3000000000001, "end": 609.48, "probability": 0.98681640625 }, { "word": "间", "start": 609.48, "end": 609.6, "probability": 1.0 }, { "word": "的", "start": 609.6, "end": 609.7, "probability": 0.99755859375 }, { "word": "时", "start": 609.7, "end": 609.82, "probability": 0.98583984375 }, { "word": "差", "start": 609.82, "end": 609.96, "probability": 0.9921875 }, { "word": "为", "start": 609.96, "end": 610.1800000000001, "probability": 0.99658203125 }, { "word": "1", "start": 610.1800000000001, "end": 610.4200000000001, "probability": 0.5634765625 }, { "word": "小", "start": 610.4200000000001, "end": 610.6, "probability": 0.99169921875 }, { "word": "时", "start": 610.6, "end": 610.9200000000001, "probability": 0.99951171875 } ], "asr_metrics": { "avg_logprob": -0.21198173308036697, "compression_ratio": 1.1597222222222223, "no_speech_prob": 2.3566499446348033e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "各位贵宾,本班季预计在当地时间下午5点30分抵达东京与天国际机场现在当地的时间为下午2点42分台北和东京之间的时差为1小时" } ], "id": 89, "review_annotation": "", "row_id": 89, "bounded_row_id": 89, "parent_row_ids": [ 89 ], "parent_row_id": 89, "packet_id": 36, "LID_evidence": [ { "region_id": 82, "probabilities": { "ja": 0.005951925675617531, "zh": 0.8540989831089973, "en": 0.08915631857234985 }, "status": "STABLE" } ], "VAD_region_ids": [ 82 ], "language_confidence": 0.8540989831089973, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 594.3000000000001, "packet_end": 612.1999999999999, "raw_text": "台北和东京之间的时差为1小时", "selected_text": "台北和东京之间的时差为1小时", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 89 ], "packet_id": 36, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "台北和东京之间的时差为1小时", "after": "台北和东京之间的时差为1小时", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Clear statement matching English context regarding time difference." } ], "language": "zh", "packet_ids": [ 36 ], "index": 89, "source_text": "台北和东京之间的时差为1小时", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Clear statement matching English context regarding time difference." }, { "start": 612.6, "end": 615.2600000000001, "text": "提醒您,您在旅途中购买的物品", "raw_asr_text": "提醒您,您在旅途中购买的物品", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 37, "words": [], "asr_metrics": { "avg_logprob": -0.11816706760324162, "compression_ratio": 1.2468085106382978, "no_speech_prob": 2.2775583502498975e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "提醒您,您在旅途中购买的物品如果有单个容器超过100毫升的液体发酵、乳液等、喷雾机等物品在转车其他航班已经回国时不得带上飞机如果有疑问请与空服人员联系若您有发烧、腹痛等症状的旅客也请与空服人员联系,谢谢" } ], "id": 90, "review_annotation": "", "row_id": 90, "bounded_row_id": 90, "parent_row_ids": [ 90 ], "parent_row_id": 90, "packet_id": 37, "LID_evidence": [ { "region_id": 83, "probabilities": { "ja": 0.012765039249526097, "zh": 0.8476619638442513, "en": 0.09781672417809587 }, "status": "STABLE" } ], "VAD_region_ids": [ 83 ], "language_confidence": 0.8476619638442513, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 611.8000000000001, "packet_end": 635.0, "raw_text": "提醒您,您在旅途中购买的物品", "selected_text": "提醒您,您在旅途中购买的物品", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 90 ], "packet_id": 37, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "提醒您,您在旅途中购买的物品", "after": "提醒您,您在旅途中购买的物品", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Matches source ASR and context; standard announcement phrasing." } ], "language": "zh", "packet_ids": [ 37 ], "index": 90, "source_text": "提醒您,您在旅途中购买的物品", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.9, "correction_reason": "Matches source ASR and context; standard announcement phrasing.", "raw_word_timestamps": [ { "word": "提", "start": 612.46, "end": 612.82, "probability": 0.88232421875 }, { "word": "醒", "start": 612.82, "end": 613.0200000000001, "probability": 0.998046875 }, { "word": "您,", "start": 613.0200000000001, "end": 613.2400000000001, "probability": 0.9365234375 }, { "word": "您", "start": 613.44, "end": 613.7600000000001, "probability": 0.9921875 }, { "word": "在", "start": 613.7600000000001, "end": 613.94, "probability": 0.998046875 }, { "word": "旅", "start": 613.94, "end": 614.1, "probability": 0.9990234375 }, { "word": "途", "start": 614.1, "end": 614.22, "probability": 0.994140625 }, { "word": "中", "start": 614.22, "end": 614.4000000000001, "probability": 0.98779296875 }, { "word": "购", "start": 614.4000000000001, "end": 614.58, "probability": 0.97802734375 }, { "word": "买", "start": 614.58, "end": 614.72, "probability": 0.9951171875 }, { "word": "的", "start": 614.72, "end": 614.86, "probability": 0.96240234375 }, { "word": "物", "start": 614.86, "end": 615.0000000000001, "probability": 0.99267578125 }, { "word": "品", "start": 615.0000000000001, "end": 615.2600000000001, "probability": 0.9970703125 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 615.2600000000001, "end": 618.0600000000001, "text": "如果有单个容器超过100毫升的液体", "raw_asr_text": "如果有单个容器超过100毫升的液体", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 37, "words": [ { "word": "如果", "start": 615.2600000000001, "end": 615.58, "probability": 0.552734375 }, { "word": "有", "start": 615.58, "end": 615.7800000000001, "probability": 0.9873046875 }, { "word": "单", "start": 615.7800000000001, "end": 615.94, "probability": 0.98193359375 }, { "word": "个", "start": 615.94, "end": 616.08, "probability": 0.97607421875 }, { "word": "容", "start": 616.08, "end": 616.2, "probability": 0.9990234375 }, { "word": "器", "start": 616.2, "end": 616.4000000000001, "probability": 0.99755859375 }, { "word": "超", "start": 616.4000000000001, "end": 616.58, "probability": 0.9970703125 }, { "word": "过", "start": 616.58, "end": 616.7800000000001, "probability": 0.998046875 }, { "word": "100", "start": 616.7800000000001, "end": 617.08, "probability": 0.93408203125 }, { "word": "毫", "start": 617.08, "end": 617.34, "probability": 0.93798828125 }, { "word": "升", "start": 617.34, "end": 617.5000000000001, "probability": 0.99462890625 }, { "word": "的", "start": 617.5000000000001, "end": 617.6400000000001, "probability": 0.98486328125 }, { "word": "液", "start": 617.6400000000001, "end": 617.82, "probability": 0.966796875 }, { "word": "体", "start": 617.82, "end": 618.0600000000001, "probability": 0.998046875 } ], "asr_metrics": { "avg_logprob": -0.11816706760324162, "compression_ratio": 1.2468085106382978, "no_speech_prob": 2.2775583502498975e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "提醒您,您在旅途中购买的物品如果有单个容器超过100毫升的液体发酵、乳液等、喷雾机等物品在转车其他航班已经回国时不得带上飞机如果有疑问请与空服人员联系若您有发烧、腹痛等症状的旅客也请与空服人员联系,谢谢" } ], "id": 91, "review_annotation": "", "row_id": 91, "bounded_row_id": 91, "parent_row_ids": [ 91 ], "parent_row_id": 91, "packet_id": 37, "LID_evidence": [ { "region_id": 83, "probabilities": { "ja": 0.012765039249526097, "zh": 0.8476619638442513, "en": 0.09781672417809587 }, "status": "STABLE" } ], "VAD_region_ids": [ 83 ], "language_confidence": 0.8476619638442513, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 611.8000000000001, "packet_end": 635.0, "raw_text": "如果有单个容器超过100毫升的液体", "selected_text": "如果有单个容器超过100毫升的液体", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 91 ], "packet_id": 37, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "如果有单个容器超过100毫升的液体", "after": "如果有单个容器超过100毫升的液体", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Matches source ASR and context; standard announcement phrasing." } ], "language": "zh", "packet_ids": [ 37 ], "index": 91, "source_text": "如果有单个容器超过100毫升的液体", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.9, "correction_reason": "Matches source ASR and context; standard announcement phrasing." }, { "start": 618.0600000000001, "end": 621.0400000000001, "text": "發膠、乳液等、噴霧機等物品", "raw_asr_text": "发酵、乳液等、喷雾机等物品", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 37, "words": [], "asr_metrics": { "avg_logprob": -0.11816706760324162, "compression_ratio": 1.2468085106382978, "no_speech_prob": 2.2775583502498975e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "提醒您,您在旅途中购买的物品如果有单个容器超过100毫升的液体发酵、乳液等、喷雾机等物品在转车其他航班已经回国时不得带上飞机如果有疑问请与空服人员联系若您有发烧、腹痛等症状的旅客也请与空服人员联系,谢谢" } ], "id": 92, "review_annotation": "", "row_id": 92, "bounded_row_id": 92, "parent_row_ids": [ 92 ], "parent_row_id": 92, "packet_id": 37, "LID_evidence": [ { "region_id": 83, "probabilities": { "ja": 0.012765039249526097, "zh": 0.8476619638442513, "en": 0.09781672417809587 }, "status": "STABLE" } ], "VAD_region_ids": [ 83 ], "language_confidence": 0.8476619638442513, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 611.8000000000001, "packet_end": 635.0, "raw_text": "发酵、乳液等、喷雾机等物品", "selected_text": "發膠、乳液等、噴霧機等物品", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 92 ], "packet_id": 37, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "发酵、乳液等、喷雾机等物品", "after": "發膠、乳液等、噴霧機等物品", "classification": "STRONG_RESTORATION", "reason": "Independent ASR confirms '發膠' (hair gel) instead of '發酵' (fermentation); '噴霧機' is likely '噴霧劑' (aerosols) misheard." } ], "language": "zh", "packet_ids": [ 37 ], "index": 92, "independent_ASR": { "text": "提醒您，您在旅途中购买的物品，如果有单个容器超过一百毫升的液体、发胶、乳液等喷雾剂等物品，在转乘其他航班及回国时不得带上飞机。如果有疑问，请与空服人员联系。若您有发烧、腹痛等症状的旅客，也请与空服人员联系。谢谢。", "chunks": [ { "text": "提醒您，您在旅途中购买的物品，如果有单个容器超过一百毫升的液体、发胶、乳液等喷雾剂等物品，在转乘其他航班及回国时不得带上飞机。如果有疑问，请与空服人员联系。若您有发烧、腹痛等症状的旅客，也请与空服人员联系。谢谢。", "start": 0.0, "end": 23.2, "chunk_index": 0, "language": "Chinese", "finish_reason": "eos", "truncated": false, "generated_tokens": 74, "max_new_tokens": 311 } ], "detected_language": "Chinese", "segments": [ { "start": 611.8000000000001, "end": 635.0000000000001, "text": "提醒您，您在旅途中购买的物品，如果有单个容器超过一百毫升的液体、发胶、乳液等喷雾剂等物品，在转乘其他航班及回国时不得带上飞机。如果有疑问，请与空服人员联系。若您有发烧、腹痛等症状的旅客，也请与空服人员联系。谢谢。" } ], "start": 611.8000000000001, "end": 635.0, "engine": "mlx-qwen3-asr", "language": "zh" }, "raw_word_timestamps": [ { "word": "发", "start": 618.0600000000001, "end": 618.7, "probability": 0.041748046875 }, { "word": "酵、", "start": 618.7, "end": 619.0400000000001, "probability": 0.896484375 }, { "word": "乳", "start": 619.4000000000001, "end": 619.5400000000001, "probability": 0.99169921875 }, { "word": "液", "start": 619.5400000000001, "end": 619.72, "probability": 0.9873046875 }, { "word": "等、", "start": 619.72, "end": 619.9000000000001, "probability": 0.94970703125 }, { "word": "喷", "start": 620.0200000000001, "end": 620.1200000000001, "probability": 0.74560546875 }, { "word": "雾", "start": 620.1200000000001, "end": 620.2600000000001, "probability": 0.9892578125 }, { "word": "机", "start": 620.2600000000001, "end": 620.4200000000001, "probability": 0.83935546875 }, { "word": "等", "start": 620.4200000000001, "end": 620.58, "probability": 0.9951171875 }, { "word": "物", "start": 620.58, "end": 620.8000000000001, "probability": 0.99560546875 }, { "word": "品", "start": 620.8000000000001, "end": 621.0400000000001, "probability": 0.99951171875 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "發膠、乳液等、噴霧機等物品", "review_status": "STRONG_RESTORATION", "classification": "STRONG_RESTORATION", "confidence": 0.9, "correction_reason": "Independent ASR confirms '發膠' (hair gel) instead of '發酵' (fermentation); '噴霧機' is likely '噴霧劑' (aerosols) misheard." }, { "start": 621.44, "end": 624.7400000000001, "text": "在轉乘其他航班及回國時不得帶上飛機", "raw_asr_text": "在转车其他航班已经回国时不得带上飞机", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 37, "words": [], "asr_metrics": { "avg_logprob": -0.11816706760324162, "compression_ratio": 1.2468085106382978, "no_speech_prob": 2.2775583502498975e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "提醒您,您在旅途中购买的物品如果有单个容器超过100毫升的液体发酵、乳液等、喷雾机等物品在转车其他航班已经回国时不得带上飞机如果有疑问请与空服人员联系若您有发烧、腹痛等症状的旅客也请与空服人员联系,谢谢" } ], "id": 93, "review_annotation": "", "row_id": 93, "bounded_row_id": 93, "parent_row_ids": [ 93 ], "parent_row_id": 93, "packet_id": 37, "LID_evidence": [ { "region_id": 83, "probabilities": { "ja": 0.012765039249526097, "zh": 0.8476619638442513, "en": 0.09781672417809587 }, "status": "STABLE" } ], "VAD_region_ids": [ 83 ], "language_confidence": 0.8476619638442513, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 611.8000000000001, "packet_end": 635.0, "raw_text": "在转车其他航班已经回国时不得带上飞机", "selected_text": "在轉乘其他航班及回國時不得帶上飛機", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 93 ], "packet_id": 37, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "在转车其他航班已经回国时不得带上飞机", "after": "在轉乘其他航班及回國時不得帶上飛機", "classification": "STRONG_RESTORATION", "reason": "Independent ASR confirms '轉乘' (transfer) and '及' (and) instead of '車' (vehicle) and '已經' (already)." } ], "language": "zh", "packet_ids": [ 37 ], "index": 93, "independent_ASR": { "text": "提醒您，您在旅途中购买的物品，如果有单个容器超过一百毫升的液体、发胶、乳液等喷雾剂等物品，在转乘其他航班及回国时不得带上飞机。如果有疑问，请与空服人员联系。若您有发烧、腹痛等症状的旅客，也请与空服人员联系。谢谢。", "chunks": [ { "text": "提醒您，您在旅途中购买的物品，如果有单个容器超过一百毫升的液体、发胶、乳液等喷雾剂等物品，在转乘其他航班及回国时不得带上飞机。如果有疑问，请与空服人员联系。若您有发烧、腹痛等症状的旅客，也请与空服人员联系。谢谢。", "start": 0.0, "end": 23.2, "chunk_index": 0, "language": "Chinese", "finish_reason": "eos", "truncated": false, "generated_tokens": 74, "max_new_tokens": 311 } ], "detected_language": "Chinese", "segments": [ { "start": 611.8000000000001, "end": 635.0000000000001, "text": "提醒您，您在旅途中购买的物品，如果有单个容器超过一百毫升的液体、发胶、乳液等喷雾剂等物品，在转乘其他航班及回国时不得带上飞机。如果有疑问，请与空服人员联系。若您有发烧、腹痛等症状的旅客，也请与空服人员联系。谢谢。" } ], "start": 611.8000000000001, "end": 635.0, "engine": "mlx-qwen3-asr", "language": "zh" }, "raw_word_timestamps": [ { "word": "在", "start": 621.44, "end": 621.8000000000001, "probability": 0.7939453125 }, { "word": "转", "start": 621.8000000000001, "end": 622.0200000000001, "probability": 0.9990234375 }, { "word": "车", "start": 622.0200000000001, "end": 622.2, "probability": 0.6943359375 }, { "word": "其", "start": 622.2, "end": 622.36, "probability": 0.96044921875 }, { "word": "他", "start": 622.36, "end": 622.5200000000001, "probability": 0.9853515625 }, { "word": "航", "start": 622.5200000000001, "end": 622.6800000000001, "probability": 0.998046875 }, { "word": "班", "start": 622.6800000000001, "end": 622.82, "probability": 0.9775390625 }, { "word": "已经", "start": 622.82, "end": 623.0200000000001, "probability": 0.452880859375 }, { "word": "回", "start": 623.0200000000001, "end": 623.1800000000001, "probability": 0.99609375 }, { "word": "国", "start": 623.1800000000001, "end": 623.4200000000001, "probability": 0.998046875 }, { "word": "时", "start": 623.4200000000001, "end": 623.6600000000001, "probability": 0.99853515625 }, { "word": "不", "start": 623.6600000000001, "end": 623.82, "probability": 0.7021484375 }, { "word": "得", "start": 623.82, "end": 623.9200000000001, "probability": 0.99609375 }, { "word": "带", "start": 623.9200000000001, "end": 624.1, "probability": 0.9658203125 }, { "word": "上", "start": 624.1, "end": 624.3000000000001, "probability": 0.9990234375 }, { "word": "飞", "start": 624.3000000000001, "end": 624.5200000000001, "probability": 0.9990234375 }, { "word": "机", "start": 624.5200000000001, "end": 624.7400000000001, "probability": 0.99951171875 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "在轉乘其他航班及回國時不得帶上飛機", "review_status": "STRONG_RESTORATION", "classification": "STRONG_RESTORATION", "confidence": 0.9, "correction_reason": "Independent ASR confirms '轉乘' (transfer) and '及' (and) instead of '車' (vehicle) and '已經' (already)." }, { "start": 624.7400000000001, "end": 627.3000000000001, "text": "如果有疑问请与空服人员联系", "raw_asr_text": "如果有疑问请与空服人员联系", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 37, "words": [ { "word": "如果", "start": 624.7400000000001, "end": 625.4000000000001, "probability": 0.96630859375 }, { "word": "有", "start": 625.4000000000001, "end": 625.58, "probability": 0.98388671875 }, { "word": "疑", "start": 625.58, "end": 625.7, "probability": 0.99609375 }, { "word": "问", "start": 625.7, "end": 625.86, "probability": 0.99755859375 }, { "word": "请", "start": 625.86, "end": 626.0600000000001, "probability": 0.6376953125 }, { "word": "与", "start": 626.0600000000001, "end": 626.22, "probability": 0.990234375 }, { "word": "空", "start": 626.22, "end": 626.44, "probability": 0.83984375 }, { "word": "服", "start": 626.44, "end": 626.5600000000001, "probability": 0.35400390625 }, { "word": "人", "start": 626.5600000000001, "end": 626.6400000000001, "probability": 0.99072265625 }, { "word": "员", "start": 626.6400000000001, "end": 626.7800000000001, "probability": 0.9990234375 }, { "word": "联", "start": 626.7800000000001, "end": 627.0200000000001, "probability": 0.9990234375 }, { "word": "系", "start": 627.0200000000001, "end": 627.3000000000001, "probability": 0.99755859375 } ], "asr_metrics": { "avg_logprob": -0.11816706760324162, "compression_ratio": 1.2468085106382978, "no_speech_prob": 2.2775583502498975e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "提醒您,您在旅途中购买的物品如果有单个容器超过100毫升的液体发酵、乳液等、喷雾机等物品在转车其他航班已经回国时不得带上飞机如果有疑问请与空服人员联系若您有发烧、腹痛等症状的旅客也请与空服人员联系,谢谢" } ], "id": 94, "review_annotation": "", "row_id": 94, "bounded_row_id": 94, "parent_row_ids": [ 94 ], "parent_row_id": 94, "packet_id": 37, "LID_evidence": [ { "region_id": 83, "probabilities": { "ja": 0.012765039249526097, "zh": 0.8476619638442513, "en": 0.09781672417809587 }, "status": "STABLE" } ], "VAD_region_ids": [ 83 ], "language_confidence": 0.8476619638442513, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 611.8000000000001, "packet_end": 635.0, "raw_text": "如果有疑问请与空服人员联系", "selected_text": "如果有疑问请与空服人员联系", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 94 ], "packet_id": 37, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "如果有疑问请与空服人员联系", "after": "如果有疑问请与空服人员联系", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Matches source ASR and context; standard announcement phrasing." } ], "language": "zh", "packet_ids": [ 37 ], "index": 94, "source_text": "如果有疑问请与空服人员联系", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.9, "correction_reason": "Matches source ASR and context; standard announcement phrasing." }, { "start": 628.3800000000001, "end": 631.2600000000001, "text": "若您有發燒、腹痛等症狀的旅客", "raw_asr_text": "若您有发烧、腹痛等症状的旅客", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 37, "words": [], "asr_metrics": { "avg_logprob": -0.11816706760324162, "compression_ratio": 1.2468085106382978, "no_speech_prob": 2.2775583502498975e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "提醒您,您在旅途中购买的物品如果有单个容器超过100毫升的液体发酵、乳液等、喷雾机等物品在转车其他航班已经回国时不得带上飞机如果有疑问请与空服人员联系若您有发烧、腹痛等症状的旅客也请与空服人员联系,谢谢" } ], "id": 95, "review_annotation": "", "row_id": 95, "bounded_row_id": 95, "parent_row_ids": [ 95 ], "parent_row_id": 95, "packet_id": 37, "LID_evidence": [ { "region_id": 84, "probabilities": { "ja": 0.009704097610665485, "zh": 0.7092060539289378, "en": 0.25185271695954725 }, "status": "STABLE" } ], "VAD_region_ids": [ 84 ], "language_confidence": 0.7092060539289378, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 611.8000000000001, "packet_end": 635.0, "raw_text": "若您有发烧、腹痛等症状的旅客", "selected_text": "若您有發燒、腹痛等症狀的旅客", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 95 ], "packet_id": 37, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "若您有发烧、腹痛等症状的旅客", "after": "若您有發燒、腹痛等症狀的旅客", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Matches independent ASR and standard announcement phrasing for passenger instructions regarding symptoms." } ], "language": "zh", "packet_ids": [ 37 ], "index": 95, "independent_ASR": { "text": "提醒您，您在旅途中购买的物品，如果有单个容器超过一百毫升的液体、发胶、乳液等喷雾剂等物品，在转乘其他航班及回国时不得带上飞机。如果有疑问，请与空服人员联系。若您有发烧、腹痛等症状的旅客，也请与空服人员联系。谢谢。", "chunks": [ { "text": "提醒您，您在旅途中购买的物品，如果有单个容器超过一百毫升的液体、发胶、乳液等喷雾剂等物品，在转乘其他航班及回国时不得带上飞机。如果有疑问，请与空服人员联系。若您有发烧、腹痛等症状的旅客，也请与空服人员联系。谢谢。", "start": 0.0, "end": 23.2, "chunk_index": 0, "language": "Chinese", "finish_reason": "eos", "truncated": false, "generated_tokens": 74, "max_new_tokens": 311 } ], "detected_language": "Chinese", "segments": [ { "start": 611.8000000000001, "end": 635.0000000000001, "text": "提醒您，您在旅途中购买的物品，如果有单个容器超过一百毫升的液体、发胶、乳液等喷雾剂等物品，在转乘其他航班及回国时不得带上飞机。如果有疑问，请与空服人员联系。若您有发烧、腹痛等症状的旅客，也请与空服人员联系。谢谢。" } ], "start": 611.8000000000001, "end": 635.0, "engine": "mlx-qwen3-asr", "language": "zh" }, "raw_word_timestamps": [ { "word": "若", "start": 628.3800000000001, "end": 628.7400000000001, "probability": 0.994140625 }, { "word": "您", "start": 628.7400000000001, "end": 628.86, "probability": 0.9375 }, { "word": "有", "start": 628.86, "end": 629.0400000000001, "probability": 0.99169921875 }, { "word": "发", "start": 629.0400000000001, "end": 629.2400000000001, "probability": 0.99853515625 }, { "word": "烧、", "start": 629.2400000000001, "end": 629.5200000000001, "probability": 0.98876953125 }, { "word": "腹", "start": 629.6800000000001, "end": 629.82, "probability": 0.9755859375 }, { "word": "痛", "start": 629.82, "end": 630.0400000000001, "probability": 0.9853515625 }, { "word": "等", "start": 630.0400000000001, "end": 630.36, "probability": 0.9912109375 }, { "word": "症", "start": 630.36, "end": 630.5600000000001, "probability": 0.9970703125 }, { "word": "状", "start": 630.5600000000001, "end": 630.7400000000001, "probability": 0.998046875 }, { "word": "的", "start": 630.7400000000001, "end": 630.8800000000001, "probability": 0.99853515625 }, { "word": "旅", "start": 630.8800000000001, "end": 631.0400000000001, "probability": 0.998046875 }, { "word": "客", "start": 631.0400000000001, "end": 631.2600000000001, "probability": 0.99951171875 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "若您有發燒、腹痛等症狀的旅客", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Matches independent ASR and standard announcement phrasing for passenger instructions regarding symptoms." }, { "start": 631.2600000000001, "end": 633.6400000000001, "text": "也请与空服人员联系,谢谢", "raw_asr_text": "也请与空服人员联系,谢谢", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 37, "words": [ { "word": "也", "start": 631.2600000000001, "end": 631.94, "probability": 0.90771484375 }, { "word": "请", "start": 631.94, "end": 632.1600000000001, "probability": 0.99951171875 }, { "word": "与", "start": 632.1600000000001, "end": 632.36, "probability": 1.0 }, { "word": "空", "start": 632.36, "end": 632.5200000000001, "probability": 0.9990234375 }, { "word": "服", "start": 632.5200000000001, "end": 632.6400000000001, "probability": 1.0 }, { "word": "人", "start": 632.6400000000001, "end": 632.7400000000001, "probability": 0.9990234375 }, { "word": "员", "start": 632.7400000000001, "end": 632.86, "probability": 1.0 }, { "word": "联", "start": 632.86, "end": 633.0400000000001, "probability": 1.0 }, { "word": "系,", "start": 633.0400000000001, "end": 633.2600000000001, "probability": 1.0 }, { "word": "谢谢", "start": 633.4200000000001, "end": 633.6400000000001, "probability": 0.99560546875 } ], "asr_metrics": { "avg_logprob": -0.11816706760324162, "compression_ratio": 1.2468085106382978, "no_speech_prob": 2.2775583502498975e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "提醒您,您在旅途中购买的物品如果有单个容器超过100毫升的液体发酵、乳液等、喷雾机等物品在转车其他航班已经回国时不得带上飞机如果有疑问请与空服人员联系若您有发烧、腹痛等症状的旅客也请与空服人员联系,谢谢" } ], "id": 96, "review_annotation": "", "row_id": 96, "bounded_row_id": 96, "parent_row_ids": [ 96 ], "parent_row_id": 96, "packet_id": 37, "LID_evidence": [ { "region_id": 84, "probabilities": { "ja": 0.009704097610665485, "zh": 0.7092060539289378, "en": 0.25185271695954725 }, "status": "STABLE" } ], "VAD_region_ids": [ 84 ], "language_confidence": 0.7092060539289378, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 611.8000000000001, "packet_end": 635.0, "raw_text": "也请与空服人员联系,谢谢", "selected_text": "也请与空服人员联系,谢谢", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 96 ], "packet_id": 37, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "也请与空服人员联系,谢谢", "after": "也请与空服人员联系,谢谢", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Matches source ASR and context; standard announcement phrasing." } ], "language": "zh", "packet_ids": [ 37 ], "index": 96, "source_text": "也请与空服人员联系,谢谢", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.9, "correction_reason": "Matches source ASR and context; standard announcement phrasing." }, { "start": 634.92, "end": 637.32, "text": "虽然现在安全带指示灯已经熄灭", "raw_asr_text": "虽然现在安全带只是灯已经熄灭", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 38, "words": [], "asr_metrics": { "avg_logprob": -0.18958166667393275, "compression_ratio": 1.0403225806451613, "no_speech_prob": 1.0432021566031402e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "虽然现在安全带只是灯已经熄灭但是由于飞行中可能突然遭遇乱流请您在旧座时人确实系好安全带" } ], "id": 97, "review_annotation": "待確認", "row_id": 97, "bounded_row_id": 97, "parent_row_ids": [ 97 ], "parent_row_id": 97, "packet_id": 38, "LID_evidence": [ { "region_id": 85, "probabilities": { "ja": 0.004539544759181273, "zh": 0.918028911246973, "en": 0.06812045947812936 }, "status": "STABLE" } ], "VAD_region_ids": [ 85 ], "language_confidence": 0.918028911246973, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 634.0, "packet_end": 644.0, "raw_text": "虽然现在安全带只是灯已经熄灭", "selected_text": "虽然现在安全带指示灯已经熄灭", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 97 ], "packet_id": 38, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "虽然现在安全带只是灯已经熄灭", "after": "虽然现在安全带指示灯已经熄灭", "classification": "BEST_SUPPORTED_WITH_REVIEW", "reason": "ASR '只是' is phonetically distinct from '指示灯'; independent ASR confirms '指示灯'. Marked 待確認 due to partial segment." } ], "language": "zh", "packet_ids": [ 38 ], "index": 97, "independent_ASR": { "text": "虽然现在安全带指示灯已经熄灭，但是由于飞行中可能突然遭遇乱流，请您在就座时仍确实系好安全带。", "chunks": [ { "text": "虽然现在安全带指示灯已经熄灭，但是由于飞行中可能突然遭遇乱流，请您在就座时仍确实系好安全带。", "start": 0.0, "end": 10.0, "chunk_index": 0, "language": "Chinese", "finish_reason": "eos", "truncated": false, "generated_tokens": 32, "max_new_tokens": 152 } ], "detected_language": "Chinese", "segments": [ { "start": 634.0, "end": 644.0, "text": "虽然现在安全带指示灯已经熄灭，但是由于飞行中可能突然遭遇乱流，请您在就座时仍确实系好安全带。" } ], "start": 634.0, "end": 644.0, "engine": "mlx-qwen3-asr", "language": "zh" }, "raw_word_timestamps": [ { "word": "虽", "start": 634.92, "end": 635.32, "probability": 0.736328125 }, { "word": "然", "start": 635.32, "end": 635.42, "probability": 1.0 }, { "word": "现在", "start": 635.42, "end": 635.64, "probability": 0.86083984375 }, { "word": "安全", "start": 635.64, "end": 635.92, "probability": 0.98779296875 }, { "word": "带", "start": 635.92, "end": 636.14, "probability": 0.9873046875 }, { "word": "只是", "start": 636.14, "end": 636.32, "probability": 0.47509765625 }, { "word": "灯", "start": 636.32, "end": 636.54, "probability": 0.958984375 }, { "word": "已经", "start": 636.54, "end": 636.8, "probability": 0.990234375 }, { "word": "熄", "start": 636.8, "end": 637.06, "probability": 0.95263671875 }, { "word": "灭", "start": 637.06, "end": 637.32, "probability": 0.99658203125 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "虽然现在安全带指示灯已经熄灭", "review_status": "BEST_SUPPORTED_WITH_REVIEW", "classification": "BEST_SUPPORTED_WITH_REVIEW", "confidence": 0.9, "correction_reason": "ASR '只是' is phonetically distinct from '指示灯'; independent ASR confirms '指示灯'. Marked 待確認 due to partial segment." }, { "start": 637.32, "end": 639.96, "text": "但是由于飞行中可能突然遭遇乱流", "raw_asr_text": "但是由于飞行中可能突然遭遇乱流", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 38, "words": [ { "word": "但是", "start": 637.32, "end": 637.82, "probability": 0.58837890625 }, { "word": "由", "start": 637.82, "end": 638.04, "probability": 0.9931640625 }, { "word": "于", "start": 638.04, "end": 638.18, "probability": 0.9990234375 }, { "word": "飞", "start": 638.18, "end": 638.3, "probability": 0.9990234375 }, { "word": "行", "start": 638.3, "end": 638.44, "probability": 1.0 }, { "word": "中", "start": 638.44, "end": 638.6, "probability": 0.9609375 }, { "word": "可能", "start": 638.6, "end": 638.84, "probability": 0.97607421875 }, { "word": "突", "start": 638.84, "end": 639.02, "probability": 0.9970703125 }, { "word": "然", "start": 639.02, "end": 639.18, "probability": 0.99951171875 }, { "word": "遭", "start": 639.18, "end": 639.36, "probability": 0.9912109375 }, { "word": "遇", "start": 639.36, "end": 639.5, "probability": 0.939453125 }, { "word": "乱", "start": 639.5, "end": 639.7, "probability": 0.6162109375 }, { "word": "流", "start": 639.7, "end": 639.96, "probability": 0.96484375 } ], "asr_metrics": { "avg_logprob": -0.18958166667393275, "compression_ratio": 1.0403225806451613, "no_speech_prob": 1.0432021566031402e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "虽然现在安全带只是灯已经熄灭但是由于飞行中可能突然遭遇乱流请您在旧座时人确实系好安全带" } ], "id": 98, "review_annotation": "", "row_id": 98, "bounded_row_id": 98, "parent_row_ids": [ 98 ], "parent_row_id": 98, "packet_id": 38, "LID_evidence": [ { "region_id": 85, "probabilities": { "ja": 0.004539544759181273, "zh": 0.918028911246973, "en": 0.06812045947812936 }, "status": "STABLE" } ], "VAD_region_ids": [ 85 ], "language_confidence": 0.918028911246973, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 634.0, "packet_end": 644.0, "raw_text": "但是由于飞行中可能突然遭遇乱流", "selected_text": "但是由于飞行中可能突然遭遇乱流", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 98 ], "packet_id": 38, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "但是由于飞行中可能突然遭遇乱流", "after": "但是由于飞行中可能突然遭遇乱流", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Matches source ASR and context; standard announcement phrasing." } ], "language": "zh", "packet_ids": [ 38 ], "index": 98, "source_text": "但是由于飞行中可能突然遭遇乱流", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.9, "correction_reason": "Matches source ASR and context; standard announcement phrasing." }, { "start": 639.96, "end": 642.8, "text": "请您在就座时仍确实系好安全带", "raw_asr_text": "请您在旧座时人确实系好安全带", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 38, "words": [], "asr_metrics": { "avg_logprob": -0.18958166667393275, "compression_ratio": 1.0403225806451613, "no_speech_prob": 1.0432021566031402e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "虽然现在安全带只是灯已经熄灭但是由于飞行中可能突然遭遇乱流请您在旧座时人确实系好安全带" } ], "id": 99, "review_annotation": "", "row_id": 99, "bounded_row_id": 99, "parent_row_ids": [ 99 ], "parent_row_id": 99, "packet_id": 38, "LID_evidence": [ { "region_id": 85, "probabilities": { "ja": 0.004539544759181273, "zh": 0.918028911246973, "en": 0.06812045947812936 }, "status": "STABLE" } ], "VAD_region_ids": [ 85 ], "language_confidence": 0.918028911246973, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 634.0, "packet_end": 644.0, "raw_text": "请您在旧座时人确实系好安全带", "selected_text": "请您在就座时仍确实系好安全带", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 99 ], "packet_id": 38, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "请您在旧座时人确实系好安全带", "after": "请您在就座时仍确实系好安全带", "classification": "CERTAIN_RESTORATION", "reason": "Independent ASR provides full sentence '就座时仍'; '旧座时人' is clear ASR nonsense for '就座时仍'." } ], "language": "zh", "packet_ids": [ 38 ], "index": 99, "independent_ASR": { "text": "虽然现在安全带指示灯已经熄灭，但是由于飞行中可能突然遭遇乱流，请您在就座时仍确实系好安全带。", "chunks": [ { "text": "虽然现在安全带指示灯已经熄灭，但是由于飞行中可能突然遭遇乱流，请您在就座时仍确实系好安全带。", "start": 0.0, "end": 10.0, "chunk_index": 0, "language": "Chinese", "finish_reason": "eos", "truncated": false, "generated_tokens": 32, "max_new_tokens": 152 } ], "detected_language": "Chinese", "segments": [ { "start": 634.0, "end": 644.0, "text": "虽然现在安全带指示灯已经熄灭，但是由于飞行中可能突然遭遇乱流，请您在就座时仍确实系好安全带。" } ], "start": 634.0, "end": 644.0, "engine": "mlx-qwen3-asr", "language": "zh" }, "raw_word_timestamps": [ { "word": "请", "start": 639.96, "end": 640.48, "probability": 0.8349609375 }, { "word": "您", "start": 640.48, "end": 640.62, "probability": 0.2998046875 }, { "word": "在", "start": 640.62, "end": 640.78, "probability": 0.74951171875 }, { "word": "旧", "start": 640.78, "end": 641.04, "probability": 0.81298828125 }, { "word": "座", "start": 641.04, "end": 641.18, "probability": 0.9501953125 }, { "word": "时", "start": 641.18, "end": 641.36, "probability": 0.75634765625 }, { "word": "人", "start": 641.36, "end": 641.52, "probability": 0.1571044921875 }, { "word": "确", "start": 641.52, "end": 641.74, "probability": 0.994140625 }, { "word": "实", "start": 641.74, "end": 641.9, "probability": 0.9990234375 }, { "word": "系", "start": 641.9, "end": 641.96, "probability": 0.7900390625 }, { "word": "好", "start": 641.96, "end": 642.16, "probability": 0.99609375 }, { "word": "安全", "start": 642.16, "end": 642.46, "probability": 0.9990234375 }, { "word": "带", "start": 642.46, "end": 642.8, "probability": 0.998046875 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "请您在就座时仍确实系好安全带", "review_status": "CERTAIN_RESTORATION", "classification": "CERTAIN_RESTORATION", "confidence": 0.95, "correction_reason": "Independent ASR provides full sentence '就座时仍'; '旧座时人' is clear ASR nonsense for '就座时仍'." }, { "start": 648.8, "end": 651.58, "text": "奥そろぎのところお連れいたします。渡辺基長です。", "raw_asr_text": "おくつそろぎのところ失礼いたします。東弁記長です。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 39, "words": [], "asr_metrics": { "avg_logprob": -0.2509052488538954, "compression_ratio": 1.3137254901960784, "no_speech_prob": 7.742891397488627e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "おくつそろぎのところ失礼いたします。東弁記長です。左手、ややこうなりますが、桜島がうっすらと水蒸気が上がっているのをご覧いただけます。" } ], "id": 100, "review_annotation": "待確認", "row_id": 100, "bounded_row_id": 100, "parent_row_ids": [ 100 ], "parent_row_id": 100, "packet_id": 39, "LID_evidence": [ { "region_id": 86, "probabilities": { "ja": 0.9989347656567892, "zh": 9.004191088024527e-05, "en": 0.00042013038182631135 }, "status": "STABLE" } ], "VAD_region_ids": [ 86 ], "language_confidence": 0.9989347656567892, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 648.0, "packet_end": 661.9, "raw_text": "おくつそろぎのところ失礼いたします。東弁記長です。", "selected_text": "奥そろぎのところお連れいたします。渡辺基長です。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 100 ], "packet_id": 39, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "おくつそろぎのところ失礼いたします。東弁記長です。", "after": "奥そろぎのところお連れいたします。渡辺基長です。", "classification": "BEST_SUPPORTED_WITH_REVIEW", "reason": "Independent ASR '奥そろぎ' and '渡辺基長' correct 'おくつ' and '東弁'. 'お連れ' likely 'お座り' or 'お越し' but '奥' is key correction." } ], "language": "ja", "packet_ids": [ 39 ], "index": 100, "independent_ASR": { "text": "奥そろぎのところお連れいたします。渡辺基長です。左手A後方になりますが桜島がうっすらと水上系が上がっているのをご覧いただけます。", "chunks": [ { "text": "奥そろぎのところお連れいたします。渡辺基長です。左手A後方になりますが桜島がうっすらと水上系が上がっているのをご覧いただけます。", "start": 0.0, "end": 13.9, "chunk_index": 0, "language": "Japanese", "finish_reason": "eos", "truncated": false, "generated_tokens": 45, "max_new_tokens": 199 } ], "detected_language": "Japanese", "segments": [ { "start": 648.0, "end": 661.9, "text": "奥そろぎのところお連れいたします。渡辺基長です。左手A後方になりますが桜島がうっすらと水上系が上がっているのをご覧いただけます。" } ], "start": 648.0, "end": 661.9, "engine": "mlx-qwen3-asr", "language": "ja" }, "raw_word_timestamps": [ { "word": "お", "start": 648.76, "end": 649.04, "probability": 0.91943359375 }, { "word": "く", "start": 649.04, "end": 649.14, "probability": 0.58935546875 }, { "word": "つ", "start": 649.14, "end": 649.22, "probability": 0.8603515625 }, { "word": "そ", "start": 649.22, "end": 649.32, "probability": 0.7451171875 }, { "word": "ろ", "start": 649.32, "end": 649.36, "probability": 0.994140625 }, { "word": "ぎ", "start": 649.36, "end": 649.42, "probability": 0.98388671875 }, { "word": "の", "start": 649.42, "end": 649.52, "probability": 0.9580078125 }, { "word": "ところ", "start": 649.52, "end": 649.82, "probability": 0.97509765625 }, { "word": "失", "start": 649.82, "end": 650.12, "probability": 0.80908203125 }, { "word": "礼", "start": 650.12, "end": 650.24, "probability": 0.9990234375 }, { "word": "いた", "start": 650.24, "end": 650.34, "probability": 0.98828125 }, { "word": "します。", "start": 650.34, "end": 650.66, "probability": 0.99951171875 }, { "word": "東", "start": 650.84, "end": 650.9, "probability": 0.6201171875 }, { "word": "弁", "start": 650.9, "end": 651.04, "probability": 0.5419921875 }, { "word": "記", "start": 651.04, "end": 651.12, "probability": 0.324462890625 }, { "word": "長", "start": 651.12, "end": 651.26, "probability": 0.6337890625 }, { "word": "です。", "start": 651.26, "end": 651.58, "probability": 1.0 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "奥そろぎのところお連れいたします。渡辺基長です。", "review_status": "BEST_SUPPORTED_WITH_REVIEW", "classification": "BEST_SUPPORTED_WITH_REVIEW", "confidence": 0.85, "correction_reason": "Independent ASR '奥そろぎ' and '渡辺基長' correct 'おくつ' and '東弁'. 'お連れ' likely 'お座り' or 'お越し' but '奥' is key correction." }, { "start": 654.0, "end": 660.78, "text": "左手、ややこうなりますが、桜島がうっすらと水蒸気が上がっているのをご覧いただけます。", "raw_asr_text": "左手、ややこうなりますが、桜島がうっすらと水蒸気が上がっているのをご覧いただけます。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 39, "words": [], "asr_metrics": { "avg_logprob": -0.2509052488538954, "compression_ratio": 1.3137254901960784, "no_speech_prob": 7.742891397488627e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "おくつそろぎのところ失礼いたします。東弁記長です。左手、ややこうなりますが、桜島がうっすらと水蒸気が上がっているのをご覧いただけます。" } ], "id": 101, "review_annotation": "", "row_id": 101, "bounded_row_id": 101, "parent_row_ids": [ 101 ], "parent_row_id": 101, "packet_id": 39, "LID_evidence": [ { "region_id": 87, "probabilities": { "ja": 0.8260551418643445, "zh": 0.0037440625715134956, "en": 0.01100290089561895 }, "status": "STABLE" } ], "VAD_region_ids": [ 87 ], "language_confidence": 0.8260551418643445, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 648.0, "packet_end": 661.9, "raw_text": "左手、ややこうなりますが、桜島がうっすらと水蒸気が上がっているのをご覧いただけます。", "selected_text": "左手、ややこうなりますが、桜島がうっすらと水蒸気が上がっているのをご覧いただけます。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 101 ], "packet_id": 39, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "左手、ややこうなりますが、桜島がうっすらと水蒸気が上がっているのをご覧いただけます。", "after": "左手、ややこうなりますが、桜島がうっすらと水蒸気が上がっているのをご覧いただけます。", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "ASR text is grammatically coherent and matches the context of a captain's scenic announcement over Kagoshima." } ], "language": "ja", "packet_ids": [ 39 ], "index": 101, "source_text": "左手、ややこうなりますが、桜島がうっすらと水蒸気が上がっているのをご覧いただけます。", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.85, "correction_reason": "ASR text is grammatically coherent and matches the context of a captain's scenic announcement over Kagoshima.", "raw_word_timestamps": [ { "word": "左", "start": 654.0, "end": 654.28, "probability": 0.99755859375 }, { "word": "手、", "start": 654.28, "end": 654.56, "probability": 0.96630859375 }, { "word": "や", "start": 654.76, "end": 654.8, "probability": 0.81787109375 }, { "word": "や", "start": 654.8, "end": 654.92, "probability": 0.99609375 }, { "word": "こう", "start": 654.92, "end": 655.08, "probability": 0.310791015625 }, { "word": "な", "start": 655.08, "end": 655.28, "probability": 0.86328125 }, { "word": "ります", "start": 655.28, "end": 655.56, "probability": 0.99951171875 }, { "word": "が、", "start": 655.56, "end": 655.72, "probability": 1.0 }, { "word": "桜", "start": 655.82, "end": 656.1, "probability": 0.9931640625 }, { "word": "島", "start": 656.1, "end": 656.36, "probability": 0.99609375 }, { "word": "が", "start": 656.36, "end": 657.62, "probability": 0.99853515625 }, { "word": "う", "start": 657.62, "end": 658.0, "probability": 0.95166015625 }, { "word": "っ", "start": 658.0, "end": 658.02, "probability": 0.7509765625 }, { "word": "す", "start": 658.02, "end": 658.1, "probability": 0.9912109375 }, { "word": "ら", "start": 658.1, "end": 658.22, "probability": 0.99951171875 }, { "word": "と", "start": 658.22, "end": 658.48, "probability": 0.95654296875 }, { "word": "水", "start": 658.48, "end": 659.2, "probability": 0.99560546875 }, { "word": "蒸", "start": 659.2, "end": 659.34, "probability": 0.919921875 }, { "word": "気", "start": 659.34, "end": 659.44, "probability": 0.98193359375 }, { "word": "が", "start": 659.44, "end": 659.56, "probability": 0.99560546875 }, { "word": "上", "start": 659.56, "end": 659.7, "probability": 0.92724609375 }, { "word": "が", "start": 659.7, "end": 659.7, "probability": 0.99755859375 }, { "word": "って", "start": 659.7, "end": 659.82, "probability": 0.9208984375 }, { "word": "いる", "start": 659.82, "end": 659.9, "probability": 0.99951171875 }, { "word": "の", "start": 659.9, "end": 659.98, "probability": 0.7705078125 }, { "word": "を", "start": 659.98, "end": 660.02, "probability": 0.97119140625 }, { "word": "ご", "start": 660.02, "end": 660.1, "probability": 0.998046875 }, { "word": "覧", "start": 660.1, "end": 660.18, "probability": 0.9990234375 }, { "word": "いただ", "start": 660.18, "end": 660.3, "probability": 0.9990234375 }, { "word": "け", "start": 660.3, "end": 660.5, "probability": 0.95263671875 }, { "word": "ます。", "start": 660.5, "end": 660.78, "probability": 1.0 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 661.72, "end": 667.34, "text": "左手下宮古の城址、九州の南端を飛行続けております", "raw_asr_text": "左手下、都の城市、九州の南端を飛行続けております", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 40, "words": [], "asr_metrics": { "avg_logprob": -0.1565604527791341, "compression_ratio": 1.2264150943396226, "no_speech_prob": 6.887508792463581e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "左手下、都の城市、九州の南端を飛行続けておりますまた間もなくしますと左手、皆様にも現在やや前方熊本の山をご覧いただけるかと思います" } ], "id": 102, "review_annotation": "待確認", "row_id": 102, "bounded_row_id": 102, "parent_row_ids": [ 102 ], "parent_row_id": 102, "packet_id": 40, "LID_evidence": [ { "region_id": 88, "probabilities": { "ja": 0.8026065868466641, "zh": 0.008116749288808822, "en": 0.009726578316820616 }, "status": "STABLE" } ], "VAD_region_ids": [ 88 ], "language_confidence": 0.8026065868466641, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 660.7, "packet_end": 678.5, "raw_text": "左手下、都の城市、九州の南端を飛行続けております", "selected_text": "左手下宮古の城址、九州の南端を飛行続けております", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 102 ], "packet_id": 40, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "左手下、都の城市、九州の南端を飛行続けております", "after": "左手下宮古の城址、九州の南端を飛行続けております", "classification": "BEST_SUPPORTED_WITH_REVIEW", "reason": "Independent ASR '宮古の城址' corrects '都の城市'. '宮古' fits geography near Kyushu/Sakurajima context better than '都'." } ], "language": "ja", "packet_ids": [ 40 ], "index": 102, "independent_ASR": { "text": "左手下宮古の城址、九州の南端を飛行続けております。またまもなくしますと左手皆様に今現在やや前方熊本の山々ご覧いただけるかと思います。", "chunks": [ { "text": "左手下宮古の城址、九州の南端を飛行続けております。またまもなくしますと左手皆様に今現在やや前方熊本の山々ご覧いただけるかと思います。", "start": 0.0, "end": 17.8, "chunk_index": 0, "language": "Japanese", "finish_reason": "eos", "truncated": false, "generated_tokens": 42, "max_new_tokens": 246 } ], "detected_language": "Japanese", "segments": [ { "start": 660.7, "end": 678.5, "text": "左手下宮古の城址、九州の南端を飛行続けております。またまもなくしますと左手皆様に今現在やや前方熊本の山々ご覧いただけるかと思います。" } ], "start": 660.7, "end": 678.5, "engine": "mlx-qwen3-asr", "language": "ja" }, "raw_word_timestamps": [ { "word": "左", "start": 661.72, "end": 662.22, "probability": 0.951171875 }, { "word": "手", "start": 662.22, "end": 662.5200000000001, "probability": 0.9990234375 }, { "word": "下、", "start": 662.5200000000001, "end": 662.86, "probability": 0.81787109375 }, { "word": "都", "start": 663.12, "end": 663.48, "probability": 0.63916015625 }, { "word": "の", "start": 663.48, "end": 663.82, "probability": 0.85009765625 }, { "word": "城", "start": 663.82, "end": 664.0600000000001, "probability": 0.88134765625 }, { "word": "市、", "start": 664.0600000000001, "end": 664.36, "probability": 0.982421875 }, { "word": "九", "start": 665.24, "end": 665.34, "probability": 0.998046875 }, { "word": "州", "start": 665.34, "end": 665.5400000000001, "probability": 1.0 }, { "word": "の", "start": 665.5400000000001, "end": 665.7, "probability": 0.9990234375 }, { "word": "南", "start": 665.7, "end": 665.9200000000001, "probability": 0.99853515625 }, { "word": "端", "start": 665.9200000000001, "end": 666.1600000000001, "probability": 0.9951171875 }, { "word": "を", "start": 666.1600000000001, "end": 666.2800000000001, "probability": 0.9990234375 }, { "word": "飛", "start": 666.2800000000001, "end": 666.38, "probability": 0.994140625 }, { "word": "行", "start": 666.38, "end": 666.5400000000001, "probability": 0.9990234375 }, { "word": "続", "start": 666.5400000000001, "end": 666.74, "probability": 0.64599609375 }, { "word": "けて", "start": 666.74, "end": 666.9200000000001, "probability": 0.9990234375 }, { "word": "お", "start": 666.9200000000001, "end": 667.0600000000001, "probability": 0.99755859375 }, { "word": "ります", "start": 667.0600000000001, "end": 667.34, "probability": 1.0 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "左手下宮古の城址、九州の南端を飛行続けております", "review_status": "BEST_SUPPORTED_WITH_REVIEW", "classification": "BEST_SUPPORTED_WITH_REVIEW", "confidence": 0.85, "correction_reason": "Independent ASR '宮古の城址' corrects '都の城市'. '宮古' fits geography near Kyushu/Sakurajima context better than '都'." }, { "start": 667.34, "end": 673.26, "text": "また、間もなくしますと左手、皆様にも現在やや前方", "raw_asr_text": "また間もなくしますと左手、皆様にも現在やや前方", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 40, "words": [ { "word": "また", "start": 667.34, "end": 668.0200000000001, "probability": 0.346435546875 }, { "word": "間", "start": 668.0200000000001, "end": 668.8000000000001, "probability": 0.329833984375 }, { "word": "も", "start": 668.8000000000001, "end": 668.9200000000001, "probability": 0.99951171875 }, { "word": "なく", "start": 668.9200000000001, "end": 669.12, "probability": 1.0 }, { "word": "します", "start": 669.12, "end": 669.4200000000001, "probability": 0.9970703125 }, { "word": "と", "start": 669.4200000000001, "end": 669.7800000000001, "probability": 1.0 }, { "word": "左", "start": 669.7800000000001, "end": 670.26, "probability": 0.708984375 }, { "word": "手、", "start": 670.26, "end": 670.72, "probability": 1.0 }, { "word": "皆", "start": 671.46, "end": 671.5400000000001, "probability": 0.89794921875 }, { "word": "様", "start": 671.5400000000001, "end": 671.7, "probability": 0.94384765625 }, { "word": "に", "start": 671.7, "end": 671.82, "probability": 0.9169921875 }, { "word": "も", "start": 671.82, "end": 671.94, "probability": 0.9541015625 }, { "word": "現在", "start": 671.94, "end": 672.2800000000001, "probability": 0.9990234375 }, { "word": "や", "start": 672.2800000000001, "end": 672.44, "probability": 0.99365234375 }, { "word": "や", "start": 672.44, "end": 672.6, "probability": 0.99951171875 }, { "word": "前", "start": 672.6, "end": 672.86, "probability": 0.99853515625 }, { "word": "方", "start": 672.86, "end": 673.26, "probability": 0.9970703125 } ], "asr_metrics": { "avg_logprob": -0.1565604527791341, "compression_ratio": 1.2264150943396226, "no_speech_prob": 6.887508792463581e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "左手下、都の城市、九州の南端を飛行続けておりますまた間もなくしますと左手、皆様にも現在やや前方熊本の山をご覧いただけるかと思います" } ], "id": 103, "review_annotation": "待確認", "row_id": 103, "bounded_row_id": 103, "parent_row_ids": [ 103 ], "parent_row_id": 103, "packet_id": 40, "LID_evidence": [ { "region_id": 88, "probabilities": { "ja": 0.8026065868466641, "zh": 0.008116749288808822, "en": 0.009726578316820616 }, "status": "STABLE" } ], "VAD_region_ids": [ 88 ], "language_confidence": 0.8026065868466641, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 660.7, "packet_end": 678.5, "raw_text": "また間もなくしますと左手、皆様にも現在やや前方", "selected_text": "また、間もなくしますと左手、皆様にも現在やや前方", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 103 ], "packet_id": 40, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "また、間もなくしますと左手、皆様にも現在やや前方", "after": "また、間もなくしますと左手、皆様にも現在やや前方", "classification": "BEST_SUPPORTED_WITH_REVIEW", "reason": "Matches independent ASR structure 'またまもなくしますと左手皆様に今現在やや前方'. Source is accurate enough." } ], "language": "ja", "deterministic_correction_applied": true, "packet_ids": [ 40 ], "index": 103, "independent_ASR": { "text": "左手下宮古の城址、九州の南端を飛行続けております。またまもなくしますと左手皆様に今現在やや前方熊本の山々ご覧いただけるかと思います。", "chunks": [ { "text": "左手下宮古の城址、九州の南端を飛行続けております。またまもなくしますと左手皆様に今現在やや前方熊本の山々ご覧いただけるかと思います。", "start": 0.0, "end": 17.8, "chunk_index": 0, "language": "Japanese", "finish_reason": "eos", "truncated": false, "generated_tokens": 42, "max_new_tokens": 246 } ], "detected_language": "Japanese", "segments": [ { "start": 660.7, "end": 678.5, "text": "左手下宮古の城址、九州の南端を飛行続けております。またまもなくしますと左手皆様に今現在やや前方熊本の山々ご覧いただけるかと思います。" } ], "start": 660.7, "end": 678.5, "engine": "mlx-qwen3-asr", "language": "ja" }, "source_text": "また、間もなくしますと左手、皆様にも現在やや前方", "review_status": "BEST_SUPPORTED_WITH_REVIEW", "classification": "BEST_SUPPORTED_WITH_REVIEW", "confidence": 0.9, "correction_reason": "Matches independent ASR structure 'またまもなくしますと左手皆様に今現在やや前方'. Source is accurate enough." }, { "start": 673.26, "end": 677.26, "text": "熊本の山をご覧いただけるかと思います", "raw_asr_text": "熊本の山をご覧いただけるかと思います", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 40, "words": [ { "word": "熊", "start": 673.26, "end": 674.1400000000001, "probability": 0.826171875 }, { "word": "本", "start": 674.1400000000001, "end": 674.4000000000001, "probability": 0.99951171875 }, { "word": "の", "start": 674.4000000000001, "end": 674.6400000000001, "probability": 0.998046875 }, { "word": "山", "start": 674.6400000000001, "end": 675.6400000000001, "probability": 0.99951171875 }, { "word": "を", "start": 675.6400000000001, "end": 676.0200000000001, "probability": 0.46484375 }, { "word": "ご", "start": 676.0200000000001, "end": 676.44, "probability": 0.9990234375 }, { "word": "覧", "start": 676.44, "end": 676.5, "probability": 0.9970703125 }, { "word": "いただ", "start": 676.5, "end": 676.6400000000001, "probability": 0.998046875 }, { "word": "ける", "start": 676.6400000000001, "end": 676.86, "probability": 0.9990234375 }, { "word": "か", "start": 676.86, "end": 676.96, "probability": 0.97900390625 }, { "word": "と思います", "start": 676.96, "end": 677.26, "probability": 0.99951171875 } ], "asr_metrics": { "avg_logprob": -0.1565604527791341, "compression_ratio": 1.2264150943396226, "no_speech_prob": 6.887508792463581e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "左手下、都の城市、九州の南端を飛行続けておりますまた間もなくしますと左手、皆様にも現在やや前方熊本の山をご覧いただけるかと思います" } ], "id": 104, "review_annotation": "", "row_id": 104, "bounded_row_id": 104, "parent_row_ids": [ 104 ], "parent_row_id": 104, "packet_id": 40, "LID_evidence": [ { "region_id": 88, "probabilities": { "ja": 0.8026065868466641, "zh": 0.008116749288808822, "en": 0.009726578316820616 }, "status": "STABLE" } ], "VAD_region_ids": [ 88 ], "language_confidence": 0.8026065868466641, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 660.7, "packet_end": 678.5, "raw_text": "熊本の山をご覧いただけるかと思います", "selected_text": "熊本の山をご覧いただけるかと思います", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 104 ], "packet_id": 40, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "熊本の山をご覧いただけるかと思います", "after": "熊本の山をご覧いただけるかと思います", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "ASR text is clear, grammatically correct, and fits the context of viewing landmarks (Kumamoto mountains)." } ], "language": "ja", "packet_ids": [ 40 ], "index": 104, "source_text": "熊本の山をご覧いただけるかと思います", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "ASR text is clear, grammatically correct, and fits the context of viewing landmarks (Kumamoto mountains)." }, { "start": 680.02, "end": 684.74, "text": "本日は上空非常に不安定な", "raw_asr_text": "本日は上空非常に不安定な", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 41, "words": [ { "word": "本", "start": 680.02, "end": 680.62, "probability": 0.84375 }, { "word": "日", "start": 680.62, "end": 680.82, "probability": 1.0 }, { "word": "は", "start": 680.82, "end": 681.12, "probability": 0.99951171875 }, { "word": "上", "start": 681.12, "end": 682.1800000000001, "probability": 0.9384765625 }, { "word": "空", "start": 682.1800000000001, "end": 682.48, "probability": 1.0 }, { "word": "非常", "start": 682.48, "end": 683.1800000000001, "probability": 0.90673828125 }, { "word": "に", "start": 683.1800000000001, "end": 683.52, "probability": 1.0 }, { "word": "不", "start": 683.52, "end": 684.02, "probability": 0.99951171875 }, { "word": "安", "start": 684.02, "end": 684.22, "probability": 1.0 }, { "word": "定", "start": 684.22, "end": 684.5, "probability": 0.99951171875 }, { "word": "な", "start": 684.5, "end": 684.74, "probability": 0.99853515625 } ], "asr_metrics": { "avg_logprob": -0.26129279817853657, "compression_ratio": 0.7659574468085106, "no_speech_prob": 8.939265126739482e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "本日は上空非常に不安定な" } ], "id": 105, "review_annotation": "", "row_id": 105, "bounded_row_id": 105, "parent_row_ids": [ 105 ], "parent_row_id": 105, "packet_id": 41, "LID_evidence": [ { "region_id": 89, "probabilities": { "ja": 0.8776240404695272, "zh": 0.001172390098872711, "en": 0.005516720346349757 }, "status": "STABLE" } ], "VAD_region_ids": [ 89 ], "language_confidence": 0.8776240404695272, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 679.1, "packet_end": 686.0, "raw_text": "本日は上空非常に不安定な", "selected_text": "本日は上空非常に不安定な", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 105 ], "packet_id": 41, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "本日は上空非常に不安定な", "after": "本日は上空非常に不安定な", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "ASR text is a valid fragment of a sentence regarding turbulence ('unstable air'), consistent with the following clause." } ], "language": "ja", "packet_ids": [ 41 ], "index": 105, "source_text": "本日は上空非常に不安定な", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.9, "correction_reason": "ASR text is a valid fragment of a sentence regarding turbulence ('unstable air'), consistent with the following clause." }, { "start": 687.84, "end": 692.96, "text": "気流によりまして通常よりも低めの高度を飛行続けております", "raw_asr_text": "気流によりまして通常よりも低めの高度を飛行続けております", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 42, "words": [ { "word": "気", "start": 687.84, "end": 688.1600000000001, "probability": 0.65625 }, { "word": "流", "start": 688.1600000000001, "end": 688.2800000000001, "probability": 0.99267578125 }, { "word": "に", "start": 688.2800000000001, "end": 688.4200000000001, "probability": 0.994140625 }, { "word": "よ", "start": 688.4200000000001, "end": 688.5, "probability": 0.99951171875 }, { "word": "り", "start": 688.5, "end": 688.58, "probability": 0.998046875 }, { "word": "ま", "start": 688.58, "end": 688.6400000000001, "probability": 0.9658203125 }, { "word": "して", "start": 688.6400000000001, "end": 688.84, "probability": 1.0 }, { "word": "通", "start": 688.84, "end": 689.5600000000001, "probability": 0.86767578125 }, { "word": "常", "start": 689.5600000000001, "end": 689.7, "probability": 0.99951171875 }, { "word": "よ", "start": 689.7, "end": 689.86, "probability": 0.99951171875 }, { "word": "り", "start": 689.86, "end": 689.9000000000001, "probability": 1.0 }, { "word": "も", "start": 689.9000000000001, "end": 690.08, "probability": 1.0 }, { "word": "低", "start": 690.08, "end": 690.8000000000001, "probability": 0.99853515625 }, { "word": "め", "start": 690.8000000000001, "end": 691.74, "probability": 0.91455078125 }, { "word": "の", "start": 691.74, "end": 691.86, "probability": 1.0 }, { "word": "高", "start": 691.86, "end": 692.0200000000001, "probability": 0.98828125 }, { "word": "度", "start": 692.0200000000001, "end": 692.1400000000001, "probability": 0.99951171875 }, { "word": "を", "start": 692.1400000000001, "end": 692.22, "probability": 0.99951171875 }, { "word": "飛", "start": 692.22, "end": 692.2800000000001, "probability": 0.990234375 }, { "word": "行", "start": 692.2800000000001, "end": 692.4000000000001, "probability": 0.9990234375 }, { "word": "続", "start": 692.4000000000001, "end": 692.5400000000001, "probability": 0.8369140625 }, { "word": "けて", "start": 692.5400000000001, "end": 692.6800000000001, "probability": 0.99951171875 }, { "word": "お", "start": 692.6800000000001, "end": 692.8000000000001, "probability": 0.9892578125 }, { "word": "ります", "start": 692.8000000000001, "end": 692.96, "probability": 1.0 } ], "asr_metrics": { "avg_logprob": -0.11054465455828973, "compression_ratio": 1.52, "no_speech_prob": 8.11125212135666e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "気流によりまして通常よりも低めの高度を飛行続けております現在23000フィート上空7000メートル上空を飛行続けていますこの先所々気流の影響を受けまして軽い揺れがあろうかと思いますできる限り安定した気流を選択しながら飛行を続けてまいりますが" } ], "id": 106, "review_annotation": "", "row_id": 106, "bounded_row_id": 106, "parent_row_ids": [ 106 ], "parent_row_id": 106, "packet_id": 42, "LID_evidence": [ { "region_id": 90, "probabilities": { "ja": 0.9858816805339995, "zh": 0.00113138255879416, "en": 0.003982445084451077 }, "status": "STABLE" } ], "VAD_region_ids": [ 90 ], "language_confidence": 0.9858816805339995, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 686.7, "packet_end": 712.6999999999999, "raw_text": "気流によりまして通常よりも低めの高度を飛行続けております", "selected_text": "気流によりまして通常よりも低めの高度を飛行続けております", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 106 ], "packet_id": 42, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "気流によりまして通常よりも低めの高度を飛行続けております", "after": "気流によりまして通常よりも低めの高度を飛行続けております", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "ASR text is clear and grammatically correct, explaining the flight path relative to air currents." } ], "language": "ja", "packet_ids": [ 42 ], "index": 106, "source_text": "気流によりまして通常よりも低めの高度を飛行続けております", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "ASR text is clear and grammatically correct, explaining the flight path relative to air currents." }, { "start": 693.4000000000001, "end": 699.0200000000001, "text": "現在23000フィート上空7000メートル上空を飛行続けています", "raw_asr_text": "現在23000フィート上空7000メートル上空を飛行続けています", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 42, "words": [ { "word": "現在", "start": 693.4000000000001, "end": 693.72, "probability": 0.7724609375 }, { "word": "23", "start": 693.72, "end": 694.0400000000001, "probability": 0.489990234375 }, { "word": "000", "start": 694.0400000000001, "end": 694.6, "probability": 0.65380859375 }, { "word": "フ", "start": 694.6, "end": 694.76, "probability": 0.92041015625 }, { "word": "ィ", "start": 694.76, "end": 694.84, "probability": 0.98876953125 }, { "word": "ート", "start": 694.84, "end": 694.96, "probability": 0.994140625 }, { "word": "上", "start": 694.96, "end": 695.32, "probability": 0.67724609375 }, { "word": "空", "start": 695.32, "end": 695.48, "probability": 1.0 }, { "word": "7", "start": 695.48, "end": 695.74, "probability": 0.93310546875 }, { "word": "000", "start": 695.74, "end": 696.0200000000001, "probability": 0.9990234375 }, { "word": "メ", "start": 696.0200000000001, "end": 696.26, "probability": 0.9228515625 }, { "word": "ート", "start": 696.26, "end": 696.38, "probability": 1.0 }, { "word": "ル", "start": 696.38, "end": 696.48, "probability": 1.0 }, { "word": "上", "start": 696.48, "end": 696.6800000000001, "probability": 0.9912109375 }, { "word": "空", "start": 696.6800000000001, "end": 696.9200000000001, "probability": 1.0 }, { "word": "を", "start": 696.9200000000001, "end": 697.1800000000001, "probability": 1.0 }, { "word": "飛", "start": 697.1800000000001, "end": 698.24, "probability": 0.99658203125 }, { "word": "行", "start": 698.24, "end": 698.36, "probability": 1.0 }, { "word": "続", "start": 698.36, "end": 698.5400000000001, "probability": 0.90771484375 }, { "word": "けて", "start": 698.5400000000001, "end": 698.74, "probability": 1.0 }, { "word": "います", "start": 698.74, "end": 699.0200000000001, "probability": 0.84912109375 } ], "asr_metrics": { "avg_logprob": -0.11054465455828973, "compression_ratio": 1.52, "no_speech_prob": 8.11125212135666e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "気流によりまして通常よりも低めの高度を飛行続けております現在23000フィート上空7000メートル上空を飛行続けていますこの先所々気流の影響を受けまして軽い揺れがあろうかと思いますできる限り安定した気流を選択しながら飛行を続けてまいりますが" } ], "id": 107, "review_annotation": "", "row_id": 107, "bounded_row_id": 107, "parent_row_ids": [ 107 ], "parent_row_id": 107, "packet_id": 42, "LID_evidence": [ { "region_id": 90, "probabilities": { "ja": 0.9858816805339995, "zh": 0.00113138255879416, "en": 0.003982445084451077 }, "status": "STABLE" } ], "VAD_region_ids": [ 90 ], "language_confidence": 0.9858816805339995, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 686.7, "packet_end": 712.6999999999999, "raw_text": "現在23000フィート上空7000メートル上空を飛行続けています", "selected_text": "現在23000フィート上空7000メートル上空を飛行続けています", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 107 ], "packet_id": 42, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "現在23000フィート上空7000メートル上空を飛行続けています", "after": "現在23000フィート上空7000メートル上空を飛行続けています", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "ASR text is clear, providing specific altitude data in feet and meters, standard for captain announcements." } ], "language": "ja", "packet_ids": [ 42 ], "index": 107, "source_text": "現在23000フィート上空7000メートル上空を飛行続けています", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "ASR text is clear, providing specific altitude data in feet and meters, standard for captain announcements." }, { "start": 700.86, "end": 705.9000000000001, "text": "この先所々気流の影響を受けまして軽い揺れがあろうかと思います", "raw_asr_text": "この先所々気流の影響を受けまして軽い揺れがあろうかと思います", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 42, "words": [ { "word": "この", "start": 700.86, "end": 701.1800000000001, "probability": 0.9990234375 }, { "word": "先", "start": 701.1800000000001, "end": 701.48, "probability": 0.9990234375 }, { "word": "所", "start": 701.48, "end": 702.1400000000001, "probability": 0.5712890625 }, { "word": "々", "start": 702.1400000000001, "end": 702.36, "probability": 0.98681640625 }, { "word": "気", "start": 702.36, "end": 703.22, "probability": 0.9697265625 }, { "word": "流", "start": 703.22, "end": 703.34, "probability": 0.99951171875 }, { "word": "の", "start": 703.34, "end": 703.46, "probability": 0.9990234375 }, { "word": "影", "start": 703.46, "end": 703.62, "probability": 1.0 }, { "word": "響", "start": 703.62, "end": 703.76, "probability": 1.0 }, { "word": "を", "start": 703.76, "end": 703.86, "probability": 0.9990234375 }, { "word": "受", "start": 703.86, "end": 703.9200000000001, "probability": 0.99951171875 }, { "word": "け", "start": 703.9200000000001, "end": 703.98, "probability": 0.9990234375 }, { "word": "ま", "start": 703.98, "end": 704.1, "probability": 0.99951171875 }, { "word": "して", "start": 704.1, "end": 704.36, "probability": 1.0 }, { "word": "軽", "start": 704.36, "end": 704.8000000000001, "probability": 0.9892578125 }, { "word": "い", "start": 704.8000000000001, "end": 704.9000000000001, "probability": 0.99658203125 }, { "word": "揺", "start": 704.9000000000001, "end": 704.96, "probability": 0.99951171875 }, { "word": "れ", "start": 704.96, "end": 705.0400000000001, "probability": 0.9990234375 }, { "word": "が", "start": 705.0400000000001, "end": 705.1800000000001, "probability": 1.0 }, { "word": "あ", "start": 705.1800000000001, "end": 705.3000000000001, "probability": 0.99658203125 }, { "word": "ろう", "start": 705.3000000000001, "end": 705.38, "probability": 1.0 }, { "word": "か", "start": 705.38, "end": 705.48, "probability": 0.99951171875 }, { "word": "と思います", "start": 705.48, "end": 705.9000000000001, "probability": 0.99853515625 } ], "asr_metrics": { "avg_logprob": -0.11054465455828973, "compression_ratio": 1.52, "no_speech_prob": 8.11125212135666e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "気流によりまして通常よりも低めの高度を飛行続けております現在23000フィート上空7000メートル上空を飛行続けていますこの先所々気流の影響を受けまして軽い揺れがあろうかと思いますできる限り安定した気流を選択しながら飛行を続けてまいりますが" } ], "id": 108, "review_annotation": "", "row_id": 108, "bounded_row_id": 108, "parent_row_ids": [ 108 ], "parent_row_id": 108, "packet_id": 42, "LID_evidence": [ { "region_id": 91, "probabilities": { "ja": 0.9085029056295753, "zh": 0.001102494090446271, "en": 0.005444506057150041 }, "status": "STABLE" } ], "VAD_region_ids": [ 91 ], "language_confidence": 0.9085029056295753, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 686.7, "packet_end": 712.6999999999999, "raw_text": "この先所々気流の影響を受けまして軽い揺れがあろうかと思います", "selected_text": "この先所々気流の影響を受けまして軽い揺れがあろうかと思います", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 108 ], "packet_id": 42, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "この先所々気流の影響を受けまして軽い揺れがあろうかと思います", "after": "この先所々気流の影響を受けまして軽い揺れがあろうかと思います", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "ASR text is clear and grammatically correct, warning passengers about potential turbulence." } ], "language": "ja", "packet_ids": [ 42 ], "index": 108, "source_text": "この先所々気流の影響を受けまして軽い揺れがあろうかと思います", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "ASR text is clear and grammatically correct, warning passengers about potential turbulence." }, { "start": 706.84, "end": 711.58, "text": "できる限り安定した気流を選択しながら飛行を続けてまいりますが", "raw_asr_text": "できる限り安定した気流を選択しながら飛行を続けてまいりますが", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 42, "words": [ { "word": "でき", "start": 706.84, "end": 707.1600000000001, "probability": 0.98828125 }, { "word": "る", "start": 707.1600000000001, "end": 707.24, "probability": 1.0 }, { "word": "限", "start": 707.24, "end": 707.4000000000001, "probability": 0.99560546875 }, { "word": "り", "start": 707.4000000000001, "end": 707.5400000000001, "probability": 0.99755859375 }, { "word": "安", "start": 707.5400000000001, "end": 707.6800000000001, "probability": 0.9990234375 }, { "word": "定", "start": 707.6800000000001, "end": 707.84, "probability": 1.0 }, { "word": "した", "start": 707.84, "end": 708.0200000000001, "probability": 0.99853515625 }, { "word": "気", "start": 708.0200000000001, "end": 708.2, "probability": 0.99951171875 }, { "word": "流", "start": 708.2, "end": 708.36, "probability": 0.99951171875 }, { "word": "を", "start": 708.36, "end": 708.6600000000001, "probability": 1.0 }, { "word": "選", "start": 708.6600000000001, "end": 709.8000000000001, "probability": 0.99560546875 }, { "word": "択", "start": 709.8000000000001, "end": 709.9200000000001, "probability": 1.0 }, { "word": "し", "start": 709.9200000000001, "end": 710.0200000000001, "probability": 0.99951171875 }, { "word": "な", "start": 710.0200000000001, "end": 710.1400000000001, "probability": 1.0 }, { "word": "が", "start": 710.1400000000001, "end": 710.2, "probability": 1.0 }, { "word": "ら", "start": 710.2, "end": 710.2800000000001, "probability": 1.0 }, { "word": "飛", "start": 710.2800000000001, "end": 710.4000000000001, "probability": 0.9951171875 }, { "word": "行", "start": 710.4000000000001, "end": 710.5600000000001, "probability": 1.0 }, { "word": "を", "start": 710.5600000000001, "end": 710.6800000000001, "probability": 0.62548828125 }, { "word": "続", "start": 710.6800000000001, "end": 710.7800000000001, "probability": 0.99951171875 }, { "word": "けて", "start": 710.7800000000001, "end": 710.98, "probability": 1.0 }, { "word": "ま", "start": 710.98, "end": 711.1400000000001, "probability": 0.99560546875 }, { "word": "い", "start": 711.1400000000001, "end": 711.1800000000001, "probability": 1.0 }, { "word": "ります", "start": 711.1800000000001, "end": 711.4200000000001, "probability": 1.0 }, { "word": "が", "start": 711.4200000000001, "end": 711.58, "probability": 0.99951171875 } ], "asr_metrics": { "avg_logprob": -0.11054465455828973, "compression_ratio": 1.52, "no_speech_prob": 8.11125212135666e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "気流によりまして通常よりも低めの高度を飛行続けております現在23000フィート上空7000メートル上空を飛行続けていますこの先所々気流の影響を受けまして軽い揺れがあろうかと思いますできる限り安定した気流を選択しながら飛行を続けてまいりますが" } ], "id": 109, "review_annotation": "", "row_id": 109, "bounded_row_id": 109, "parent_row_ids": [ 109 ], "parent_row_id": 109, "packet_id": 42, "LID_evidence": [ { "region_id": 92, "probabilities": { "ja": 0.7558318336804708, "zh": 0.003927147150534438, "en": 0.017484394668523844 }, "status": "STABLE" } ], "VAD_region_ids": [ 92 ], "language_confidence": 0.7558318336804708, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 686.7, "packet_end": 712.6999999999999, "raw_text": "できる限り安定した気流を選択しながら飛行を続けてまいりますが", "selected_text": "できる限り安定した気流を選択しながら飛行を続けてまいりますが", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 109 ], "packet_id": 42, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "できる限り安定した気流を選択しながら飛行を続けてまいりますが", "after": "できる限り安定した気流を選択しながら飛行を続けてまいりますが", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "ASR text is clear and grammatically correct, describing the flight crew's effort to maintain stability." } ], "language": "ja", "packet_ids": [ 42 ], "index": 109, "source_text": "できる限り安定した気流を選択しながら飛行を続けてまいりますが", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "ASR text is clear and grammatically correct, describing the flight crew's effort to maintain stability." }, { "start": 713.5400000000001, "end": 717.1400000000001, "text": "揺れましても安全上心配ありませんのでどうぞご安心ください", "raw_asr_text": "揺れましても安全上昇がございませんのでどうぞご安心ください", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 43, "words": [], "asr_metrics": { "avg_logprob": -0.18058316972520616, "compression_ratio": 1.391705069124424, "no_speech_prob": 8.051961875032987e-12 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "揺れましても安全上昇がございませんのでどうぞご安心ください飛行時間はあと1時間10分ほどで羽田空港に着陸いたします天候は雨気温は7度でございます少し冷え込んでいるようでございますゲート到着は帝国の17時30分からありがとうございました" } ], "id": 110, "review_annotation": "", "row_id": 110, "bounded_row_id": 110, "parent_row_ids": [ 110 ], "parent_row_id": 110, "packet_id": 43, "LID_evidence": [ { "region_id": 93, "probabilities": { "ja": 0.7642218196764589, "zh": 0.004419975531345699, "en": 0.01997883198200725 }, "status": "STABLE" } ], "VAD_region_ids": [ 93 ], "language_confidence": 0.7642218196764589, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 711.9000000000001, "packet_end": 738.9, "raw_text": "揺れましても安全上昇がございませんのでどうぞご安心ください", "selected_text": "揺れましても安全上心配ありませんのでどうぞご安心ください", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 110 ], "packet_id": 43, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "揺れましても安全上昇がございませんのでどうぞご安心ください", "after": "揺れましても安全上心配ありませんのでどうぞご安心ください", "classification": "CERTAIN_RESTORATION", "reason": "Independent ASR confirms '安全上心配ありません' which corrects the ASR nonsense '安全上昇'." } ], "language": "ja", "packet_ids": [ 43 ], "index": 110, "independent_ASR": { "text": "揺れましても安全上心配ありませんのでどうぞご安心ください。飛行時間はあと1時間10分ほどで羽田空港に着陸いたします。天候は雨、気温は7度でございます。少し冷え込んでるようでございます。ゲート到着は定刻の17時30分から。", "chunks": [ { "text": "揺れましても安全上心配ありませんのでどうぞご安心ください。飛行時間はあと1時間10分ほどで羽田空港に着陸いたします。天候は雨、気温は7度でございます。少し冷え込んでるようでございます。ゲート到着は定刻の17時30分から。", "start": 0.0, "end": 27.0, "chunk_index": 0, "language": "Japanese", "finish_reason": "eos", "truncated": false, "generated_tokens": 75, "max_new_tokens": 356 } ], "detected_language": "Japanese", "segments": [ { "start": 711.9000000000001, "end": 738.9000000000001, "text": "揺れましても安全上心配ありませんのでどうぞご安心ください。飛行時間はあと1時間10分ほどで羽田空港に着陸いたします。天候は雨、気温は7度でございます。少し冷え込んでるようでございます。ゲート到着は定刻の17時30分から。" } ], "start": 711.9000000000001, "end": 738.9, "engine": "mlx-qwen3-asr", "language": "ja" }, "raw_word_timestamps": [ { "word": "揺", "start": 713.5400000000001, "end": 713.94, "probability": 0.86083984375 }, { "word": "れ", "start": 713.94, "end": 714.0200000000001, "probability": 0.99169921875 }, { "word": "ま", "start": 714.0200000000001, "end": 714.1400000000001, "probability": 0.99267578125 }, { "word": "して", "start": 714.1400000000001, "end": 714.2600000000001, "probability": 1.0 }, { "word": "も", "start": 714.2600000000001, "end": 714.4200000000001, "probability": 0.99951171875 }, { "word": "安全", "start": 714.4200000000001, "end": 714.7200000000001, "probability": 0.94384765625 }, { "word": "上", "start": 714.7200000000001, "end": 714.9200000000001, "probability": 0.91357421875 }, { "word": "昇", "start": 714.9200000000001, "end": 715.08, "probability": 0.845703125 }, { "word": "が", "start": 715.08, "end": 715.2600000000001, "probability": 0.70654296875 }, { "word": "ござ", "start": 715.2600000000001, "end": 715.3600000000001, "probability": 0.41064453125 }, { "word": "い", "start": 715.3600000000001, "end": 715.4800000000001, "probability": 0.998046875 }, { "word": "ません", "start": 715.4800000000001, "end": 715.6800000000001, "probability": 0.99951171875 }, { "word": "ので", "start": 715.6800000000001, "end": 715.96, "probability": 0.99951171875 }, { "word": "どう", "start": 715.96, "end": 716.32, "probability": 0.59375 }, { "word": "ぞ", "start": 716.32, "end": 716.46, "probability": 1.0 }, { "word": "ご", "start": 716.46, "end": 716.58, "probability": 0.564453125 }, { "word": "安", "start": 716.58, "end": 716.6600000000001, "probability": 0.94287109375 }, { "word": "心", "start": 716.6600000000001, "end": 716.7400000000001, "probability": 0.82275390625 }, { "word": "ください", "start": 716.7400000000001, "end": 717.1400000000001, "probability": 0.99462890625 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "揺れましても安全上心配ありませんのでどうぞご安心ください", "review_status": "CERTAIN_RESTORATION", "classification": "CERTAIN_RESTORATION", "confidence": 0.95, "correction_reason": "Independent ASR confirms '安全上心配ありません' which corrects the ASR nonsense '安全上昇'." }, { "start": 718.4800000000001, "end": 727.32, "text": "飛行時間はあと1時間10分ほどで羽田空港に着陸いたします", "raw_asr_text": "飛行時間はあと1時間10分ほどで羽田空港に着陸いたします", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 43, "words": [ { "word": "飛", "start": 718.4800000000001, "end": 718.8800000000001, "probability": 0.480224609375 }, { "word": "行", "start": 718.8800000000001, "end": 719.0400000000001, "probability": 0.99951171875 }, { "word": "時間", "start": 719.0400000000001, "end": 719.3600000000001, "probability": 1.0 }, { "word": "は", "start": 719.3600000000001, "end": 719.7, "probability": 0.99951171875 }, { "word": "あと", "start": 719.7, "end": 721.7400000000001, "probability": 0.95751953125 }, { "word": "1", "start": 721.7400000000001, "end": 722.2200000000001, "probability": 0.9990234375 }, { "word": "時間", "start": 722.2200000000001, "end": 722.6800000000001, "probability": 1.0 }, { "word": "10", "start": 722.6800000000001, "end": 723.3800000000001, "probability": 0.99169921875 }, { "word": "分", "start": 723.3800000000001, "end": 723.7600000000001, "probability": 1.0 }, { "word": "ほど", "start": 723.7600000000001, "end": 724.08, "probability": 0.99365234375 }, { "word": "で", "start": 724.08, "end": 724.4200000000001, "probability": 0.99951171875 }, { "word": "羽", "start": 724.4200000000001, "end": 725.2, "probability": 0.99609375 }, { "word": "田", "start": 725.2, "end": 725.3800000000001, "probability": 0.9951171875 }, { "word": "空", "start": 725.3800000000001, "end": 725.6000000000001, "probability": 1.0 }, { "word": "港", "start": 725.6000000000001, "end": 725.8800000000001, "probability": 0.99951171875 }, { "word": "に", "start": 725.8800000000001, "end": 726.08, "probability": 0.99951171875 }, { "word": "着", "start": 726.08, "end": 726.3600000000001, "probability": 1.0 }, { "word": "陸", "start": 726.3600000000001, "end": 726.6400000000001, "probability": 0.99951171875 }, { "word": "いた", "start": 726.6400000000001, "end": 726.9000000000001, "probability": 0.9951171875 }, { "word": "します", "start": 726.9000000000001, "end": 727.32, "probability": 1.0 } ], "asr_metrics": { "avg_logprob": -0.18058316972520616, "compression_ratio": 1.391705069124424, "no_speech_prob": 8.051961875032987e-12 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "揺れましても安全上昇がございませんのでどうぞご安心ください飛行時間はあと1時間10分ほどで羽田空港に着陸いたします天候は雨気温は7度でございます少し冷え込んでいるようでございますゲート到着は帝国の17時30分からありがとうございました" } ], "id": 111, "review_annotation": "", "row_id": 111, "bounded_row_id": 111, "parent_row_ids": [ 111 ], "parent_row_id": 111, "packet_id": 43, "LID_evidence": [ { "region_id": 94, "probabilities": { "ja": 0.7428536417428404, "zh": 0.0016472404458909295, "en": 0.002676546006114222 }, "status": "UNCERTAIN" }, { "region_id": 95, "probabilities": { "ja": 0.9296292186404268, "zh": 0.0005908338406394857, "en": 0.0014571409109824648 }, "status": "STABLE" } ], "VAD_region_ids": [ 94, 95 ], "language_confidence": 0.9296292186404268, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 711.9000000000001, "packet_end": 738.9, "raw_text": "飛行時間はあと1時間10分ほどで羽田空港に着陸いたします", "selected_text": "飛行時間はあと1時間10分ほどで羽田空港に着陸いたします", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 111 ], "packet_id": 43, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "飛行時間はあと1時間10分ほどで羽田空港に着陸いたします", "after": "飛行時間はあと1時間10分ほどで羽田空港に着陸いたします", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "文脈と日本語の表現として自然であり、ASR結果も正確です。" } ], "language": "ja", "packet_ids": [ 43 ], "index": 111, "source_text": "飛行時間はあと1時間10分ほどで羽田空港に着陸いたします", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "文脈と日本語の表現として自然であり、ASR結果も正確です。" }, { "start": 728.9200000000001, "end": 732.8000000000001, "text": "天候は雨気温は7度でございます少し冷え込んでいるようでございます", "raw_asr_text": "天候は雨気温は7度でございます少し冷え込んでいるようでございます", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 43, "words": [ { "word": "天", "start": 728.9200000000001, "end": 729.32, "probability": 0.99462890625 }, { "word": "候", "start": 729.32, "end": 729.5200000000001, "probability": 0.98583984375 }, { "word": "は", "start": 729.5200000000001, "end": 729.7, "probability": 0.78369140625 }, { "word": "雨", "start": 729.7, "end": 729.82, "probability": 0.99951171875 }, { "word": "気", "start": 729.82, "end": 730.0600000000001, "probability": 0.298828125 }, { "word": "温", "start": 730.0600000000001, "end": 730.1600000000001, "probability": 1.0 }, { "word": "は", "start": 730.1600000000001, "end": 730.3000000000001, "probability": 0.99951171875 }, { "word": "7", "start": 730.3000000000001, "end": 730.46, "probability": 0.99658203125 }, { "word": "度", "start": 730.46, "end": 730.6400000000001, "probability": 0.99755859375 }, { "word": "で", "start": 730.6400000000001, "end": 730.7200000000001, "probability": 0.99853515625 }, { "word": "ございます", "start": 730.7200000000001, "end": 731.1400000000001, "probability": 0.99951171875 }, { "word": "少", "start": 731.1400000000001, "end": 731.5600000000001, "probability": 0.9931640625 }, { "word": "し", "start": 731.5600000000001, "end": 731.7, "probability": 1.0 }, { "word": "冷", "start": 731.7, "end": 731.82, "probability": 0.99951171875 }, { "word": "え", "start": 731.82, "end": 731.9200000000001, "probability": 0.998046875 }, { "word": "込", "start": 731.9200000000001, "end": 732.0200000000001, "probability": 0.9951171875 }, { "word": "んで", "start": 732.0200000000001, "end": 732.1400000000001, "probability": 0.88671875 }, { "word": "いる", "start": 732.1400000000001, "end": 732.2, "probability": 0.95556640625 }, { "word": "よう", "start": 732.2, "end": 732.3000000000001, "probability": 0.974609375 }, { "word": "で", "start": 732.3000000000001, "end": 732.3600000000001, "probability": 0.9765625 }, { "word": "ございます", "start": 732.3600000000001, "end": 732.8000000000001, "probability": 1.0 } ], "asr_metrics": { "avg_logprob": -0.18058316972520616, "compression_ratio": 1.391705069124424, "no_speech_prob": 8.051961875032987e-12 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "揺れましても安全上昇がございませんのでどうぞご安心ください飛行時間はあと1時間10分ほどで羽田空港に着陸いたします天候は雨気温は7度でございます少し冷え込んでいるようでございますゲート到着は帝国の17時30分からありがとうございました" } ], "id": 112, "review_annotation": "", "row_id": 112, "bounded_row_id": 112, "parent_row_ids": [ 112 ], "parent_row_id": 112, "packet_id": 43, "LID_evidence": [ { "region_id": 96, "probabilities": { "ja": 0.9964584112167358, "zh": 0.00030077478877501564, "en": 0.001402941334526986 }, "status": "STABLE" } ], "VAD_region_ids": [ 96 ], "language_confidence": 0.9964584112167358, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 711.9000000000001, "packet_end": 738.9, "raw_text": "天候は雨気温は7度でございます少し冷え込んでいるようでございます", "selected_text": "天候は雨気温は7度でございます少し冷え込んでいるようでございます", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 112 ], "packet_id": 43, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "天候は雨気温は7度でございます少し冷え込んでいるようでございます", "after": "天候は雨気温は7度でございます少し冷え込んでいるようでございます", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "文脈と日本語の表現として自然であり、ASR結果も正確です。" } ], "language": "ja", "packet_ids": [ 43 ], "index": 112, "source_text": "天候は雨気温は7度でございます少し冷え込んでいるようでございます", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "文脈と日本語の表現として自然であり、ASR結果も正確です。" }, { "start": 734.2600000000001, "end": 737.6800000000001, "text": "ゲート到着は定刻の17時30分から", "raw_asr_text": "ゲート到着は帝国の17時30分から", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 43, "words": [], "asr_metrics": { "avg_logprob": -0.18058316972520616, "compression_ratio": 1.391705069124424, "no_speech_prob": 8.051961875032987e-12 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "揺れましても安全上昇がございませんのでどうぞご安心ください飛行時間はあと1時間10分ほどで羽田空港に着陸いたします天候は雨気温は7度でございます少し冷え込んでいるようでございますゲート到着は帝国の17時30分からありがとうございました" } ], "id": 113, "review_annotation": "", "row_id": 113, "bounded_row_id": 113, "parent_row_ids": [ 113 ], "parent_row_id": 113, "packet_id": 43, "LID_evidence": [ { "region_id": 97, "probabilities": { "ja": 0.9975026369094848, "zh": 0.0002131863686372526, "en": 0.0010625800292473286 }, "status": "STABLE" } ], "VAD_region_ids": [ 97 ], "language_confidence": 0.9975026369094848, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 711.9000000000001, "packet_end": 738.9, "raw_text": "ゲート到着は帝国の17時30分から", "selected_text": "ゲート到着は定刻の17時30分から", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 113 ], "packet_id": 43, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "ゲート到着は帝国の17時30分から", "after": "ゲート到着は定刻の17時30分から", "classification": "CERTAIN_RESTORATION", "reason": "Independent ASR confirms '定刻' (scheduled time) correcting the nonsense '帝国'." } ], "language": "ja", "packet_ids": [ 43 ], "index": 113, "independent_ASR": { "text": "揺れましても安全上心配ありませんのでどうぞご安心ください。飛行時間はあと1時間10分ほどで羽田空港に着陸いたします。天候は雨、気温は7度でございます。少し冷え込んでるようでございます。ゲート到着は定刻の17時30分から。", "chunks": [ { "text": "揺れましても安全上心配ありませんのでどうぞご安心ください。飛行時間はあと1時間10分ほどで羽田空港に着陸いたします。天候は雨、気温は7度でございます。少し冷え込んでるようでございます。ゲート到着は定刻の17時30分から。", "start": 0.0, "end": 27.0, "chunk_index": 0, "language": "Japanese", "finish_reason": "eos", "truncated": false, "generated_tokens": 75, "max_new_tokens": 356 } ], "detected_language": "Japanese", "segments": [ { "start": 711.9000000000001, "end": 738.9000000000001, "text": "揺れましても安全上心配ありませんのでどうぞご安心ください。飛行時間はあと1時間10分ほどで羽田空港に着陸いたします。天候は雨、気温は7度でございます。少し冷え込んでるようでございます。ゲート到着は定刻の17時30分から。" } ], "start": 711.9000000000001, "end": 738.9, "engine": "mlx-qwen3-asr", "language": "ja" }, "raw_word_timestamps": [ { "word": "ゲ", "start": 734.2600000000001, "end": 734.6600000000001, "probability": 0.998046875 }, { "word": "ート", "start": 734.6600000000001, "end": 734.82, "probability": 0.99951171875 }, { "word": "到", "start": 734.82, "end": 734.96, "probability": 0.99951171875 }, { "word": "着", "start": 734.96, "end": 735.1600000000001, "probability": 1.0 }, { "word": "は", "start": 735.1600000000001, "end": 735.3800000000001, "probability": 1.0 }, { "word": "帝", "start": 735.3800000000001, "end": 735.6000000000001, "probability": 0.74267578125 }, { "word": "国", "start": 735.6000000000001, "end": 735.82, "probability": 0.99755859375 }, { "word": "の", "start": 735.82, "end": 736.08, "probability": 0.97998046875 }, { "word": "17", "start": 736.08, "end": 736.6200000000001, "probability": 0.99755859375 }, { "word": "時", "start": 736.6200000000001, "end": 736.82, "probability": 0.9990234375 }, { "word": "30", "start": 736.82, "end": 737.08, "probability": 0.99951171875 }, { "word": "分", "start": 737.08, "end": 737.4200000000001, "probability": 1.0 }, { "word": "から", "start": 737.4200000000001, "end": 737.6800000000001, "probability": 1.0 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "ゲート到着は定刻の17時30分から", "review_status": "CERTAIN_RESTORATION", "classification": "CERTAIN_RESTORATION", "confidence": 0.95, "correction_reason": "Independent ASR confirms '定刻' (scheduled time) correcting the nonsense '帝国'." }, { "start": 737.6800000000001, "end": 737.7800000000001, "text": "ありがとうございました", "raw_asr_text": "ありがとうございました", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 43, "words": [ { "word": "ありがとうございました", "start": 737.6800000000001, "end": 737.7, "probability": 0.338134765625 } ], "asr_metrics": { "avg_logprob": -0.6233577728271484, "compression_ratio": 0.75, "no_speech_prob": 1.251016512604508e-10 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "揺れましても安全上昇がございませんのでどうぞご安心ください飛行時間はあと1時間10分ほどで羽田空港に着陸いたします天候は雨気温は7度でございます少し冷え込んでいるようでございますゲート到着は帝国の17時30分からありがとうございました" } ], "id": 114, "review_annotation": "", "row_id": 114, "bounded_row_id": 114, "parent_row_ids": [ 114 ], "parent_row_id": 114, "packet_id": 43, "LID_evidence": [ { "region_id": 97, "probabilities": { "ja": 0.9975026369094848, "zh": 0.0002131863686372526, "en": 0.0010625800292473286 }, "status": "STABLE" } ], "VAD_region_ids": [ 97 ], "language_confidence": 0.9975026369094848, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 711.9000000000001, "packet_end": 738.9, "raw_text": "ありがとうございました", "selected_text": "ありがとうございました", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 114 ], "packet_id": 43, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "ありがとうございました", "after": "ありがとうございました", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard closing phrase, consistent with context and independent ASR." } ], "language": "ja", "packet_ids": [ 43 ], "index": 114, "independent_ASR": { "text": "揺れましても安全上心配ありませんのでどうぞご安心ください。飛行時間はあと1時間10分ほどで羽田空港に着陸いたします。天候は雨、気温は7度でございます。少し冷え込んでるようでございます。ゲート到着は定刻の17時30分から。", "chunks": [ { "text": "揺れましても安全上心配ありませんのでどうぞご安心ください。飛行時間はあと1時間10分ほどで羽田空港に着陸いたします。天候は雨、気温は7度でございます。少し冷え込んでるようでございます。ゲート到着は定刻の17時30分から。", "start": 0.0, "end": 27.0, "chunk_index": 0, "language": "Japanese", "finish_reason": "eos", "truncated": false, "generated_tokens": 75, "max_new_tokens": 356 } ], "detected_language": "Japanese", "segments": [ { "start": 711.9000000000001, "end": 738.9000000000001, "text": "揺れましても安全上心配ありませんのでどうぞご安心ください。飛行時間はあと1時間10分ほどで羽田空港に着陸いたします。天候は雨、気温は7度でございます。少し冷え込んでるようでございます。ゲート到着は定刻の17時30分から。" } ], "start": 711.9000000000001, "end": 738.9, "engine": "mlx-qwen3-asr", "language": "ja" }, "source_text": "ありがとうございました", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.9, "correction_reason": "Standard closing phrase, consistent with context and independent ASR.", "timing_adjustment": { "old_start": 737.6800000000001, "old_end": 737.7, "new_start": 737.6800000000001, "new_end": 737.7800000000001, "method": "PCM voiced envelope, conservative100ms precision", "transaction": "COMMITTED" } }, { "start": 739.6200000000001, "end": 740.8800000000001, "text": "15分ほど早く", "raw_asr_text": "15分ほど早く", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 44, "words": [ { "word": "15", "start": 739.6200000000001, "end": 739.96, "probability": 0.50390625 }, { "word": "分", "start": 739.96, "end": 740.2, "probability": 0.99951171875 }, { "word": "ほど", "start": 740.2, "end": 740.4200000000001, "probability": 0.994140625 }, { "word": "早", "start": 740.4200000000001, "end": 740.6400000000001, "probability": 0.9970703125 }, { "word": "く", "start": 740.6400000000001, "end": 740.8800000000001, "probability": 1.0 } ], "asr_metrics": { "avg_logprob": -0.15537501218026145, "compression_ratio": 1.2587412587412588, "no_speech_prob": 8.63798269024052e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "15分ほど早く本日は17時15分の到着予定しておりますごゆっくりと過ごしてくださいご登場を御礼申し上げますありがとうございました" } ], "id": 115, "review_annotation": "", "row_id": 115, "bounded_row_id": 115, "parent_row_ids": [ 115 ], "parent_row_id": 115, "packet_id": 44, "LID_evidence": [ { "region_id": 98, "probabilities": { "ja": 0.997868025302887, "zh": 0.0001580273936269805, "en": 0.0009285957959946245 }, "status": "STABLE" } ], "VAD_region_ids": [ 98 ], "language_confidence": 0.997868025302887, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 738.3000000000001, "packet_end": 750.9, "raw_text": "15分ほど早く", "selected_text": "15分ほど早く", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 115 ], "packet_id": 44, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "15分ほど早く", "after": "15分ほど早く", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "文脈と日本語の表現として自然であり、ASR結果も正確です。" } ], "language": "ja", "packet_ids": [ 44 ], "index": 115, "source_text": "15分ほど早く", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "文脈と日本語の表現として自然であり、ASR結果も正確です。" }, { "start": 740.8800000000001, "end": 744.7, "text": "本日は17時15分の到着予定しております", "raw_asr_text": "本日は17時15分の到着予定しております", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 44, "words": [ { "word": "本", "start": 740.8800000000001, "end": 741.5600000000001, "probability": 0.2998046875 }, { "word": "日", "start": 741.5600000000001, "end": 741.7400000000001, "probability": 1.0 }, { "word": "は", "start": 741.7400000000001, "end": 741.96, "probability": 1.0 }, { "word": "17", "start": 741.96, "end": 742.58, "probability": 0.99365234375 }, { "word": "時", "start": 742.58, "end": 742.84, "probability": 0.99951171875 }, { "word": "15", "start": 742.84, "end": 743.2400000000001, "probability": 0.99658203125 }, { "word": "分", "start": 743.2400000000001, "end": 743.46, "probability": 1.0 }, { "word": "の", "start": 743.46, "end": 743.58, "probability": 0.9990234375 }, { "word": "到", "start": 743.58, "end": 743.7400000000001, "probability": 0.99853515625 }, { "word": "着", "start": 743.7400000000001, "end": 743.9200000000001, "probability": 0.99951171875 }, { "word": "予", "start": 743.9200000000001, "end": 744.1, "probability": 0.76708984375 }, { "word": "定", "start": 744.1, "end": 744.2, "probability": 0.99951171875 }, { "word": "して", "start": 744.2, "end": 744.34, "probability": 0.9990234375 }, { "word": "お", "start": 744.34, "end": 744.44, "probability": 0.99951171875 }, { "word": "ります", "start": 744.44, "end": 744.7, "probability": 1.0 } ], "asr_metrics": { "avg_logprob": -0.15537501218026145, "compression_ratio": 1.2587412587412588, "no_speech_prob": 8.63798269024052e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "15分ほど早く本日は17時15分の到着予定しておりますごゆっくりと過ごしてくださいご登場を御礼申し上げますありがとうございました" } ], "id": 116, "review_annotation": "", "row_id": 116, "bounded_row_id": 116, "parent_row_ids": [ 116 ], "parent_row_id": 116, "packet_id": 44, "LID_evidence": [ { "region_id": 98, "probabilities": { "ja": 0.997868025302887, "zh": 0.0001580273936269805, "en": 0.0009285957959946245 }, "status": "STABLE" }, { "region_id": 99, "probabilities": { "ja": 0.9975592344999313, "zh": 0.00018214839747088263, "en": 0.0010334799299016595 }, "status": "STABLE" } ], "VAD_region_ids": [ 98, 99 ], "language_confidence": 0.997868025302887, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 738.3000000000001, "packet_end": 750.9, "raw_text": "本日は17時15分の到着予定しております", "selected_text": "本日は17時15分の到着予定しております", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 116 ], "packet_id": 44, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "本日は17時15分の到着予定しております", "after": "本日は17時15分の到着予定しております", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "文脈と日本語の表現として自然であり、ASR結果も正確です。" } ], "language": "ja", "packet_ids": [ 44 ], "index": 116, "source_text": "本日は17時15分の到着予定しております", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "文脈と日本語の表現として自然であり、ASR結果も正確です。" }, { "start": 746.0600000000001, "end": 747.3800000000001, "text": "ごゆっくりと過ごしてください", "raw_asr_text": "ごゆっくりと過ごしてください", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 44, "words": [ { "word": "ご", "start": 746.0600000000001, "end": 746.4000000000001, "probability": 0.96728515625 }, { "word": "ゆ", "start": 746.4000000000001, "end": 746.48, "probability": 0.77099609375 }, { "word": "っ", "start": 746.48, "end": 746.5200000000001, "probability": 0.99755859375 }, { "word": "く", "start": 746.5200000000001, "end": 746.58, "probability": 0.998046875 }, { "word": "り", "start": 746.58, "end": 746.6400000000001, "probability": 0.99951171875 }, { "word": "と", "start": 746.6400000000001, "end": 746.7600000000001, "probability": 0.99609375 }, { "word": "過", "start": 746.7600000000001, "end": 746.9200000000001, "probability": 0.60400390625 }, { "word": "ご", "start": 746.9200000000001, "end": 746.98, "probability": 0.99951171875 }, { "word": "して", "start": 746.98, "end": 747.0600000000001, "probability": 0.52490234375 }, { "word": "ください", "start": 747.0600000000001, "end": 747.3800000000001, "probability": 0.99462890625 } ], "asr_metrics": { "avg_logprob": -0.15537501218026145, "compression_ratio": 1.2587412587412588, "no_speech_prob": 8.63798269024052e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "15分ほど早く本日は17時15分の到着予定しておりますごゆっくりと過ごしてくださいご登場を御礼申し上げますありがとうございました" } ], "id": 117, "review_annotation": "", "row_id": 117, "bounded_row_id": 117, "parent_row_ids": [ 117 ], "parent_row_id": 117, "packet_id": 44, "LID_evidence": [ { "region_id": 100, "probabilities": { "ja": 0.9963335514068603, "zh": 0.00027735913172364233, "en": 0.0017710320884361864 }, "status": "STABLE" } ], "VAD_region_ids": [ 100 ], "language_confidence": 0.9963335514068603, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 738.3000000000001, "packet_end": 750.9, "raw_text": "ごゆっくりと過ごしてください", "selected_text": "ごゆっくりと過ごしてください", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 117 ], "packet_id": 44, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "ごゆっくりと過ごしてください", "after": "ごゆっくりと過ごしてください", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "文脈と日本語の表現として自然であり、ASR結果も正確です。" } ], "language": "ja", "packet_ids": [ 44 ], "index": 117, "source_text": "ごゆっくりと過ごしてください", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "文脈と日本語の表現として自然であり、ASR結果も正確です。" }, { "start": 747.3800000000001, "end": 748.9000000000001, "text": "ご搭乗を御礼申し上げます", "raw_asr_text": "ご登場を御礼申し上げます", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 44, "words": [], "asr_metrics": { "avg_logprob": -0.15537501218026145, "compression_ratio": 1.2587412587412588, "no_speech_prob": 8.63798269024052e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "15分ほど早く本日は17時15分の到着予定しておりますごゆっくりと過ごしてくださいご登場を御礼申し上げますありがとうございました" } ], "id": 118, "review_annotation": "待確認", "row_id": 118, "bounded_row_id": 118, "parent_row_ids": [ 118 ], "parent_row_id": 118, "packet_id": 44, "LID_evidence": [ { "region_id": 100, "probabilities": { "ja": 0.9963335514068603, "zh": 0.00027735913172364233, "en": 0.0017710320884361864 }, "status": "STABLE" } ], "VAD_region_ids": [ 100 ], "language_confidence": 0.9963335514068603, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 738.3000000000001, "packet_end": 750.9, "raw_text": "ご登場を御礼申し上げます", "selected_text": "ご搭乗を御礼申し上げます", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 118 ], "packet_id": 44, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "ご登場を御礼申し上げます", "after": "ご搭乗を御礼申し上げます", "classification": "BEST_SUPPORTED_WITH_REVIEW", "reason": "Independent ASR 'ご登場お念' is nonsense; 'ご搭乗' (boarding) is the standard phrase fitting the context." } ], "language": "ja", "packet_ids": [ 44 ], "index": 118, "independent_ASR": { "text": "15分ほど早く本日は17時15分の到着予定しております。ごゆっくりと過ごしてください。ご登場お念申し上げます。ありがとうございました。", "chunks": [ { "text": "15分ほど早く本日は17時15分の到着予定しております。ごゆっくりと過ごしてください。ご登場お念申し上げます。ありがとうございました。", "start": 0.0, "end": 12.6, "chunk_index": 0, "language": "Japanese", "finish_reason": "eos", "truncated": false, "generated_tokens": 38, "max_new_tokens": 184 } ], "detected_language": "Japanese", "segments": [ { "start": 738.3000000000001, "end": 750.9000000000001, "text": "15分ほど早く本日は17時15分の到着予定しております。ごゆっくりと過ごしてください。ご登場お念申し上げます。ありがとうございました。" } ], "start": 738.3000000000001, "end": 750.9, "engine": "mlx-qwen3-asr", "language": "ja" }, "raw_word_timestamps": [ { "word": "ご", "start": 747.3800000000001, "end": 747.5600000000001, "probability": 0.95849609375 }, { "word": "登", "start": 747.5600000000001, "end": 747.72, "probability": 0.76220703125 }, { "word": "場", "start": 747.72, "end": 747.8800000000001, "probability": 0.99951171875 }, { "word": "を", "start": 747.8800000000001, "end": 748.0400000000001, "probability": 0.8046875 }, { "word": "御", "start": 748.0400000000001, "end": 748.1400000000001, "probability": 0.787109375 }, { "word": "礼", "start": 748.1400000000001, "end": 748.2400000000001, "probability": 0.98779296875 }, { "word": "申", "start": 748.2400000000001, "end": 748.48, "probability": 0.998046875 }, { "word": "し", "start": 748.48, "end": 748.5200000000001, "probability": 1.0 }, { "word": "上", "start": 748.5200000000001, "end": 748.6200000000001, "probability": 1.0 }, { "word": "げ", "start": 748.6200000000001, "end": 748.6800000000001, "probability": 1.0 }, { "word": "ます", "start": 748.6800000000001, "end": 748.9000000000001, "probability": 0.99951171875 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "ご搭乗を御礼申し上げます", "review_status": "BEST_SUPPORTED_WITH_REVIEW", "classification": "BEST_SUPPORTED_WITH_REVIEW", "confidence": 0.85, "correction_reason": "Independent ASR 'ご登場お念' is nonsense; 'ご搭乗' (boarding) is the standard phrase fitting the context." }, { "start": 748.9000000000001, "end": 749.1800000000001, "text": "ありがとうございました", "raw_asr_text": "ありがとうございました", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 44, "words": [ { "word": "ありがとうございました", "start": 748.9000000000001, "end": 749.1800000000001, "probability": 0.97998046875 } ], "asr_metrics": { "avg_logprob": -0.15537501218026145, "compression_ratio": 1.2587412587412588, "no_speech_prob": 8.63798269024052e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "15分ほど早く本日は17時15分の到着予定しておりますごゆっくりと過ごしてくださいご登場を御礼申し上げますありがとうございました" } ], "id": 119, "review_annotation": "", "row_id": 119, "bounded_row_id": 119, "parent_row_ids": [ 119 ], "parent_row_id": 119, "packet_id": 44, "LID_evidence": [ { "region_id": 100, "probabilities": { "ja": 0.9963335514068603, "zh": 0.00027735913172364233, "en": 0.0017710320884361864 }, "status": "STABLE" } ], "VAD_region_ids": [ 100 ], "language_confidence": 0.9963335514068603, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 738.3000000000001, "packet_end": 750.9, "raw_text": "ありがとうございました", "selected_text": "ありがとうございました", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 119 ], "packet_id": 44, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "ありがとうございました", "after": "ありがとうございました", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard closing phrase, consistent with context and independent ASR." } ], "language": "ja", "packet_ids": [ 44 ], "index": 119, "independent_ASR": { "text": "15分ほど早く本日は17時15分の到着予定しております。ごゆっくりと過ごしてください。ご登場お念申し上げます。ありがとうございました。", "chunks": [ { "text": "15分ほど早く本日は17時15分の到着予定しております。ごゆっくりと過ごしてください。ご登場お念申し上げます。ありがとうございました。", "start": 0.0, "end": 12.6, "chunk_index": 0, "language": "Japanese", "finish_reason": "eos", "truncated": false, "generated_tokens": 38, "max_new_tokens": 184 } ], "detected_language": "Japanese", "segments": [ { "start": 738.3000000000001, "end": 750.9000000000001, "text": "15分ほど早く本日は17時15分の到着予定しております。ごゆっくりと過ごしてください。ご登場お念申し上げます。ありがとうございました。" } ], "start": 738.3000000000001, "end": 750.9, "engine": "mlx-qwen3-asr", "language": "ja" }, "source_text": "ありがとうございました", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.9, "correction_reason": "Standard closing phrase, consistent with context and independent ASR." }, { "start": 752.3, "end": 775.12, "text": "This is captain speaking. We are cruising 23,000, it's about 7,000 meters and ground speed at 570 knots. And one for Tokyo. I expect a nice ride. Another 30 minutes, and I expect the light or bumpier when we start to descend for Tokyo. So when you have a seat, please fasten your seat belt.", "raw_asr_text": "This is captain speaking. We are cruising 23,000, it's about 7,000 meters and ground speed at 570 knots and one for Tokyo. I expect a nice ride, another 30 minutes and I expect the light or bumpy air when we start to descend for Tokyo. So when you have us seated, please fasten the seatbelt.", "source_language": "en", "source_lang": "en", "asr_packet_id": 45, "words": [], "asr_metrics": { "avg_logprob": -0.36463721593221027, "compression_ratio": 1.4696969696969697, "no_speech_prob": 3.9120595651809253e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "en", "text": "This is captain speaking. We are cruising 23,000, it's about 7,000 meters and ground speed at 570 knots and one for Tokyo. I expect a nice ride, another 30 minutes and I expect the light or bumpy air when we start to descend for Tokyo. So when you have us seated, please fasten the seatbelt." } ], "id": 120, "review_annotation": "待確認", "row_id": 120, "bounded_row_id": 120, "parent_row_ids": [ 120 ], "parent_row_id": 120, "packet_id": 45, "LID_evidence": [ { "region_id": 101, "probabilities": { "ja": 0.0038048520606632033, "zh": 0.00492313257806624, "en": 0.7548835873603821 }, "status": "STABLE" }, { "region_id": 102, "probabilities": { "ja": 0.003968817268340739, "zh": 0.002912524895852103, "en": 0.8779259183827568 }, "status": "STABLE" }, { "region_id": 103, "probabilities": { "ja": 0.003348633417044766, "zh": 0.0032788476150017233, "en": 0.7478885788470506 }, "status": "STABLE" }, { "region_id": 104, "probabilities": { "ja": 0.003312759411831697, "zh": 0.002894943735251824, "en": 0.9007103443145752 }, "status": "STABLE" }, { "region_id": 105, "probabilities": { "ja": 0.007095108869382077, "zh": 0.007216336313932616, "en": 0.8898684249983894 }, "status": "STABLE" }, { "region_id": 106, "probabilities": { "ja": 0.008746513786415258, "zh": 0.005893121762912617, "en": 0.8942347963651022 }, "status": "STABLE" } ], "VAD_region_ids": [ 101, 102, 103, 104, 105, 106 ], "language_confidence": 0.9007103443145752, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 751.5, "packet_end": 776.1999999999999, "raw_text": "This is captain speaking. We are cruising 23,000, it's about 7,000 meters and ground speed at 570 knots and one for Tokyo. I expect a nice ride, another 30 minutes and I expect the light or bumpy air when we start to descend for Tokyo. So when you have us seated, please fasten the seatbelt.", "selected_text": "This is captain speaking. We are cruising 23,000, it's about 7,000 meters and ground speed at 570 knots. And one for Tokyo. I expect a nice ride. Another 30 minutes, and I expect the light or bumpier when we start to descend for Tokyo. So when you have a seat, please fasten your seat belt.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 120 ], "packet_id": 45, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "This is captain speaking. We are cruising 23,000, it's about 7,000 meters and ground speed at 570 knots and one for Tokyo. I expect a nice ride, another 30 minutes and I expect the light or bumpy air when we start to descend for Tokyo. So when you have us seated, please fasten the seatbelt.", "after": "This is captain speaking. We are cruising 23,000, it's about 7,000 meters and ground speed at 570 knots. And one for Tokyo. I expect a nice ride. Another 30 minutes, and I expect the light or bumpier when we start to descend for Tokyo. So when you have a seat, please fasten your seat belt.", "classification": "BEST_SUPPORTED_WITH_REVIEW", "reason": "Independent ASR provides better grammar ('ground speed of', 'bumpier', 'have a seat') correcting source nonsense." } ], "language": "en", "packet_ids": [ 45 ], "index": 120, "independent_ASR": { "text": "This is Captain speaking. We are cruising 23,000. It's about 7,000 meters and ground speed of 570 knots. And one for Tokyo. I expect a nice ride. Another 30 minutes, and I expect the light or bumpier when we start to descend for Tokyo. So when you have a seat, please fasten your seat belt. The", "chunks": [ { "text": "This is Captain speaking. We are cruising 23,000. It's about 7,000 meters and ground speed of 570 knots. And one for Tokyo. I expect a nice ride. Another 30 minutes, and I expect the light or bumpier when we start to descend for Tokyo. So when you have a seat, please fasten your seat belt. The", "start": 0.0, "end": 24.7, "chunk_index": 0, "language": "English", "finish_reason": "eos", "truncated": false, "generated_tokens": 84, "max_new_tokens": 329 } ], "detected_language": "English", "segments": [ { "start": 751.5, "end": 776.2, "text": "This is Captain speaking. We are cruising 23,000. It's about 7,000 meters and ground speed of 570 knots. And one for Tokyo. I expect a nice ride. Another 30 minutes, and I expect the light or bumpier when we start to descend for Tokyo. So when you have a seat, please fasten your seat belt. The" } ], "start": 751.5, "end": 776.1999999999999, "engine": "mlx-qwen3-asr", "language": "en" }, "raw_word_timestamps": [ { "word": " This", "start": 752.3, "end": 752.86, "probability": 0.67431640625 }, { "word": " is", "start": 752.86, "end": 753.04, "probability": 0.98193359375 }, { "word": " captain", "start": 753.04, "end": 753.42, "probability": 0.466796875 }, { "word": " speaking.", "start": 753.42, "end": 754.08, "probability": 0.9794921875 }, { "word": " We", "start": 755.5, "end": 756.06, "probability": 0.9599609375 }, { "word": " are", "start": 756.06, "end": 756.18, "probability": 0.75341796875 }, { "word": " cruising", "start": 756.18, "end": 756.46, "probability": 0.9658203125 }, { "word": " 23", "start": 756.46, "end": 756.98, "probability": 0.5869140625 }, { "word": ",000,", "start": 756.98, "end": 757.32, "probability": 0.19921875 }, { "word": " it's", "start": 757.32, "end": 757.54, "probability": 0.68017578125 }, { "word": " about", "start": 757.54, "end": 757.7, "probability": 0.9833984375 }, { "word": " 7", "start": 757.7, "end": 757.96, "probability": 0.6689453125 }, { "word": ",000", "start": 757.96, "end": 758.22, "probability": 0.8310546875 }, { "word": " meters", "start": 758.22, "end": 758.5, "probability": 0.67041015625 }, { "word": " and", "start": 758.5, "end": 758.76, "probability": 0.60986328125 }, { "word": " ground", "start": 758.76, "end": 758.96, "probability": 0.84814453125 }, { "word": " speed", "start": 758.96, "end": 759.2, "probability": 0.97216796875 }, { "word": " at", "start": 759.2, "end": 759.42, "probability": 0.398193359375 }, { "word": " 570", "start": 759.42, "end": 759.96, "probability": 0.830078125 }, { "word": " knots", "start": 759.96, "end": 760.6, "probability": 0.97021484375 }, { "word": " and", "start": 760.6, "end": 760.94, "probability": 0.5634765625 }, { "word": " one", "start": 760.94, "end": 761.08, "probability": 0.552734375 }, { "word": " for", "start": 761.08, "end": 761.24, "probability": 0.9755859375 }, { "word": " Tokyo.", "start": 761.24, "end": 761.58, "probability": 0.96826171875 }, { "word": " I", "start": 762.0, "end": 762.54, "probability": 0.93798828125 }, { "word": " expect", "start": 762.54, "end": 762.82, "probability": 0.9765625 }, { "word": " a", "start": 762.82, "end": 763.0, "probability": 0.88037109375 }, { "word": " nice", "start": 763.0, "end": 763.14, "probability": 0.99658203125 }, { "word": " ride,", "start": 763.14, "end": 763.42, "probability": 0.994140625 }, { "word": " another", "start": 763.52, "end": 763.76, "probability": 0.98046875 }, { "word": " 30", "start": 763.76, "end": 764.04, "probability": 0.92626953125 }, { "word": " minutes", "start": 764.04, "end": 764.36, "probability": 0.9814453125 }, { "word": " and", "start": 764.36, "end": 764.78, "probability": 0.50390625 }, { "word": " I", "start": 764.78, "end": 765.0, "probability": 0.9677734375 }, { "word": " expect", "start": 765.0, "end": 765.74, "probability": 0.9658203125 }, { "word": " the", "start": 765.74, "end": 766.0, "probability": 0.5966796875 }, { "word": " light", "start": 766.0, "end": 766.82, "probability": 0.95068359375 }, { "word": " or", "start": 766.82, "end": 768.6, "probability": 0.81005859375 }, { "word": " bumpy", "start": 768.6, "end": 768.9, "probability": 0.419921875 }, { "word": " air", "start": 768.9, "end": 769.24, "probability": 0.9716796875 }, { "word": " when", "start": 769.24, "end": 770.26, "probability": 0.86572265625 }, { "word": " we", "start": 770.26, "end": 770.38, "probability": 0.98681640625 }, { "word": " start", "start": 770.38, "end": 770.6, "probability": 0.97265625 }, { "word": " to", "start": 770.6, "end": 770.74, "probability": 0.90380859375 }, { "word": " descend", "start": 770.74, "end": 771.08, "probability": 0.71826171875 }, { "word": " for", "start": 771.08, "end": 771.88, "probability": 0.81689453125 }, { "word": " Tokyo.", "start": 771.88, "end": 772.3, "probability": 0.99609375 }, { "word": " So", "start": 772.72, "end": 773.18, "probability": 0.9599609375 }, { "word": " when", "start": 773.18, "end": 773.36, "probability": 0.859375 }, { "word": " you", "start": 773.36, "end": 773.42, "probability": 0.978515625 }, { "word": " have", "start": 773.42, "end": 773.58, "probability": 0.990234375 }, { "word": " us", "start": 773.58, "end": 773.76, "probability": 0.65625 }, { "word": " seated,", "start": 773.76, "end": 773.92, "probability": 0.7802734375 }, { "word": " please", "start": 774.1, "end": 774.28, "probability": 0.99658203125 }, { "word": " fasten", "start": 774.28, "end": 774.68, "probability": 0.7509765625 }, { "word": " the", "start": 774.68, "end": 774.82, "probability": 0.7470703125 }, { "word": " seatbelt.", "start": 774.82, "end": 775.12, "probability": 0.7138671875 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "This is captain speaking. We are cruising 23,000, it's about 7,000 meters and ground speed at 570 knots. And one for Tokyo. I expect a nice ride. Another 30 minutes, and I expect the light or bumpier when we start to descend for Tokyo. So when you have a seat, please fasten your seat belt.", "review_status": "BEST_SUPPORTED_WITH_REVIEW", "classification": "BEST_SUPPORTED_WITH_REVIEW", "confidence": 0.8, "correction_reason": "Independent ASR provides better grammar ('ground speed of', 'bumpier', 'have a seat') correcting source nonsense." }, { "start": 775.9, "end": 779.98, "text": "The Tokyo's weather is raining condition and 7 degrees Celsius.", "raw_asr_text": "The Tokyo's weather is raining condition and 7 degrees Celsius.", "source_language": "en", "source_lang": "en", "asr_packet_id": 46, "words": [], "asr_metrics": { "avg_logprob": -0.43021488189697266, "compression_ratio": 1.3020134228187918, "no_speech_prob": 4.029899677848725e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "en", "text": "The Tokyo's weather is raining condition and 7 degrees Celsius. Landing time will be 5.05, a litigate and a 5.15 local time, and 15 minutes before the schedule. Once again, thank you for voting." }, { "language": "ja", "text": "東京の暖かい気持ちが降りると、7C5.0515分後、15分前の時間が降りると、また、ご覧いただきありがとうございます" } ], "id": 121, "review_annotation": "", "row_id": 121, "bounded_row_id": 121, "parent_row_ids": [ 121 ], "parent_row_id": 121, "packet_id": 46, "LID_evidence": [ { "region_id": 107, "probabilities": { "ja": 0.0059045136709196065, "zh": 0.0035293761399001456, "en": 0.8583102512818116 }, "status": "STABLE" } ], "VAD_region_ids": [ 107 ], "language_confidence": 0.8583102512818116, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 775.1, "packet_end": 794.5, "raw_text": "The Tokyo's weather is raining condition and 7 degrees Celsius.", "selected_text": "The Tokyo's weather is raining condition and 7 degrees Celsius.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 121 ], "packet_id": 46, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "The Tokyo's weather is raining condition and 7 degrees Celsius.", "after": "The Tokyo's weather is raining condition and 7 degrees Celsius.", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Grammatically awkward but matches ASR and context of non-native speaker announcement." } ], "language": "en", "packet_ids": [ 46 ], "index": 121, "source_text": "The Tokyo's weather is raining condition and 7 degrees Celsius.", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.9, "correction_reason": "Grammatically awkward but matches ASR and context of non-native speaker announcement.", "raw_word_timestamps": [ { "word": " The", "start": 775.1, "end": 776.14, "probability": 0.3134765625 }, { "word": " Tokyo's", "start": 776.14, "end": 776.72, "probability": 0.6103515625 }, { "word": " weather", "start": 776.72, "end": 777.5, "probability": 0.490234375 }, { "word": " is", "start": 777.5, "end": 777.8000000000001, "probability": 0.98388671875 }, { "word": " raining", "start": 777.8000000000001, "end": 778.32, "probability": 0.693359375 }, { "word": " condition", "start": 778.32, "end": 778.7, "probability": 0.537109375 }, { "word": " and", "start": 778.7, "end": 779.0, "probability": 0.517578125 }, { "word": " 7", "start": 779.0, "end": 779.3000000000001, "probability": 0.583984375 }, { "word": " degrees", "start": 779.3000000000001, "end": 779.5400000000001, "probability": 0.7646484375 }, { "word": " Celsius.", "start": 779.5400000000001, "end": 779.98, "probability": 0.8701171875 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 780.62, "end": 791.58, "text": "The Tokyo's weather is raining condition and seven degrees Celsius. Landing time will be at five o' five. A little gate in the five fifteen local time, in fifteen minutes before the schedule. Once again, thank you for voting.", "raw_asr_text": "Landing time will be 5.05, a litigate and a 5.15 local time, and 15 minutes before the schedule.", "source_language": "en", "source_lang": "en", "asr_packet_id": 46, "words": [], "asr_metrics": { "avg_logprob": -0.43021488189697266, "compression_ratio": 1.3020134228187918, "no_speech_prob": 4.029899677848725e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "en", "text": "The Tokyo's weather is raining condition and 7 degrees Celsius. Landing time will be 5.05, a litigate and a 5.15 local time, and 15 minutes before the schedule. Once again, thank you for voting." }, { "language": "ja", "text": "東京の暖かい気持ちが降りると、7C5.0515分後、15分前の時間が降りると、また、ご覧いただきありがとうございます" } ], "id": 122, "review_annotation": "待確認", "row_id": 122, "bounded_row_id": 122, "parent_row_ids": [ 122 ], "parent_row_id": 122, "packet_id": 46, "LID_evidence": [ { "region_id": 108, "probabilities": { "ja": 0.0031644588258738318, "zh": 0.002574666985310614, "en": 0.7970379417141279 }, "status": "STABLE" }, { "region_id": 109, "probabilities": { "ja": 0.002829285920597613, "zh": 0.0027302881935611367, "en": 0.8537545800209045 }, "status": "UNCERTAIN" }, { "region_id": 110, "probabilities": { "ja": 0, "zh": 0, "en": 0 }, "status": "UNCERTAIN" }, { "region_id": 111, "probabilities": { "ja": 0.007197921207989566, "zh": 0.004876003775279969, "en": 0.8481639921665192 }, "status": "STABLE" }, { "region_id": 112, "probabilities": { "ja": 0.005936246049324316, "zh": 0.0036894073044615133, "en": 0.8585144111088344 }, "status": "STABLE" } ], "VAD_region_ids": [ 108, 109, 110, 111, 112 ], "language_confidence": 0.8585144111088344, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 775.1, "packet_end": 794.5, "raw_text": "Landing time will be 5.05, a litigate and a 5.15 local time, and 15 minutes before the schedule.", "selected_text": "The Tokyo's weather is raining condition and seven degrees Celsius. Landing time will be at five o' five. A little gate in the five fifteen local time, in fifteen minutes before the schedule. Once again, thank you for voting.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 122 ], "packet_id": 46, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "Landing time will be 5.05, a litigate and a 5.15 local time, and 15 minutes before the schedule.", "after": "The Tokyo's weather is raining condition and seven degrees Celsius. Landing time will be at five o' five. A little gate in the five fifteen local time, in fifteen minutes before the schedule. Once again, thank you for voting.", "classification": "BEST_SUPPORTED_WITH_REVIEW", "reason": "Independent ASR provides clearer phonetic structure for time and weather, correcting source nonsense." } ], "language": "en", "packet_ids": [ 46 ], "index": 122, "independent_ASR": { "text": "The Tokyo's weather is raining condition and seven degrees Celsius. Landing time will be at five o' five. A little gate in the five fifteen local time, in fifteen minutes before the schedule. Once again, thank you for voting.", "chunks": [ { "text": "The Tokyo's weather is raining condition and seven degrees Celsius. Landing time will be at five o' five. A little gate in the five fifteen local time, in fifteen minutes before the schedule. Once again, thank you for voting.", "start": 0.0, "end": 19.4, "chunk_index": 0, "language": "English", "finish_reason": "eos", "truncated": false, "generated_tokens": 47, "max_new_tokens": 265 } ], "detected_language": "English", "segments": [ { "start": 775.1, "end": 794.5, "text": "The Tokyo's weather is raining condition and seven degrees Celsius. Landing time will be at five o' five. A little gate in the five fifteen local time, in fifteen minutes before the schedule. Once again, thank you for voting." } ], "start": 775.1, "end": 794.5, "engine": "mlx-qwen3-asr", "language": "en" }, "raw_word_timestamps": [ { "word": " Landing", "start": 780.62, "end": 781.22, "probability": 0.329833984375 }, { "word": " time", "start": 781.22, "end": 781.44, "probability": 0.9814453125 }, { "word": " will", "start": 781.44, "end": 781.62, "probability": 0.95703125 }, { "word": " be", "start": 781.62, "end": 781.8000000000001, "probability": 0.9951171875 }, { "word": " 5", "start": 781.8000000000001, "end": 783.14, "probability": 0.64599609375 }, { "word": ".05,", "start": 783.14, "end": 785.3000000000001, "probability": 0.58935546875 }, { "word": " a", "start": 786.24, "end": 787.0400000000001, "probability": 0.296630859375 }, { "word": " litigate", "start": 787.0400000000001, "end": 787.5, "probability": 0.56103515625 }, { "word": " and", "start": 787.5, "end": 787.64, "probability": 0.5556640625 }, { "word": " a", "start": 787.64, "end": 787.78, "probability": 0.57861328125 }, { "word": " 5", "start": 787.78, "end": 787.9, "probability": 0.96337890625 }, { "word": ".15", "start": 787.9, "end": 788.28, "probability": 0.90087890625 }, { "word": " local", "start": 788.28, "end": 788.6, "probability": 0.955078125 }, { "word": " time,", "start": 788.6, "end": 788.96, "probability": 0.99609375 }, { "word": " and", "start": 789.14, "end": 789.2, "probability": 0.435546875 }, { "word": " 15", "start": 789.2, "end": 789.5, "probability": 0.85400390625 }, { "word": " minutes", "start": 789.5, "end": 789.9, "probability": 0.99072265625 }, { "word": " before", "start": 789.9, "end": 791.12, "probability": 0.98828125 }, { "word": " the", "start": 791.12, "end": 791.3000000000001, "probability": 0.9951171875 }, { "word": " schedule.", "start": 791.3000000000001, "end": 791.58, "probability": 0.9892578125 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "The Tokyo's weather is raining condition and seven degrees Celsius. Landing time will be at five o' five. A little gate in the five fifteen local time, in fifteen minutes before the schedule. Once again, thank you for voting.", "review_status": "BEST_SUPPORTED_WITH_REVIEW", "classification": "BEST_SUPPORTED_WITH_REVIEW", "confidence": 0.8, "correction_reason": "Independent ASR provides clearer phonetic structure for time and weather, correcting source nonsense." }, { "start": 791.96, "end": 793.34, "text": "Once again, thank you for voting.", "raw_asr_text": "Once again, thank you for voting.", "source_language": "en", "source_lang": "en", "asr_packet_id": 46, "words": [ { "word": " Once", "start": 791.96, "end": 792.4200000000001, "probability": 0.99609375 }, { "word": " again,", "start": 792.4200000000001, "end": 792.66, "probability": 0.99853515625 }, { "word": " thank", "start": 792.8000000000001, "end": 792.9, "probability": 0.8798828125 }, { "word": " you", "start": 792.9, "end": 793.02, "probability": 0.9873046875 }, { "word": " for", "start": 793.02, "end": 793.1, "probability": 0.994140625 }, { "word": " voting.", "start": 793.1, "end": 793.34, "probability": 0.9609375 } ], "asr_metrics": { "avg_logprob": -0.43021488189697266, "compression_ratio": 1.3020134228187918, "no_speech_prob": 4.029899677848725e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "en", "text": "The Tokyo's weather is raining condition and 7 degrees Celsius. Landing time will be 5.05, a litigate and a 5.15 local time, and 15 minutes before the schedule. Once again, thank you for voting." }, { "language": "ja", "text": "東京の暖かい気持ちが降りると、7C5.0515分後、15分前の時間が降りると、また、ご覧いただきありがとうございます" } ], "id": 123, "review_annotation": "待確認", "row_id": 123, "bounded_row_id": 123, "parent_row_ids": [ 123 ], "parent_row_id": 123, "packet_id": 46, "LID_evidence": [ { "region_id": 112, "probabilities": { "ja": 0.005936246049324316, "zh": 0.0036894073044615133, "en": 0.8585144111088344 }, "status": "STABLE" }, { "region_id": 113, "probabilities": { "ja": 0.006711511930916458, "zh": 0.0033313423045910895, "en": 0.9273172169923782 }, "status": "STABLE" } ], "VAD_region_ids": [ 112, 113 ], "language_confidence": 0.9273172169923782, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 775.1, "packet_end": 794.5, "raw_text": "Once again, thank you for voting.", "selected_text": "Once again, thank you for voting.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 123 ], "packet_id": 46, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "Once again, thank you for voting.", "after": "Once again, thank you for voting.", "classification": "BEST_SUPPORTED_WITH_REVIEW", "reason": "Independent ASR confirms 'voting' (likely 'boarding' misheard), preserving source wording as supported." } ], "language": "en", "packet_ids": [ 46 ], "index": 123, "independent_ASR": { "text": "The Tokyo's weather is raining condition and seven degrees Celsius. Landing time will be at five o' five. A little gate in the five fifteen local time, in fifteen minutes before the schedule. Once again, thank you for voting.", "chunks": [ { "text": "The Tokyo's weather is raining condition and seven degrees Celsius. Landing time will be at five o' five. A little gate in the five fifteen local time, in fifteen minutes before the schedule. Once again, thank you for voting.", "start": 0.0, "end": 19.4, "chunk_index": 0, "language": "English", "finish_reason": "eos", "truncated": false, "generated_tokens": 47, "max_new_tokens": 265 } ], "detected_language": "English", "segments": [ { "start": 775.1, "end": 794.5, "text": "The Tokyo's weather is raining condition and seven degrees Celsius. Landing time will be at five o' five. A little gate in the five fifteen local time, in fifteen minutes before the schedule. Once again, thank you for voting." } ], "start": 775.1, "end": 794.5, "engine": "mlx-qwen3-asr", "language": "en" }, "source_text": "Once again, thank you for voting.", "review_status": "BEST_SUPPORTED_WITH_REVIEW", "classification": "BEST_SUPPORTED_WITH_REVIEW", "confidence": 0.85, "correction_reason": "Independent ASR confirms 'voting' (likely 'boarding' misheard), preserving source wording as supported." }, { "start": 794.9, "end": 800.28, "text": "你好，我是志全。感谢乘坐本次航班", "raw_asr_text": "你好,我是吉田,感谢重座本次航班。", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 47, "words": [], "asr_metrics": { "avg_logprob": -0.5655769525572311, "compression_ratio": 0.9326923076923077, "no_speech_prob": 1.587830561478043e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "zh", "text": "你好,我是吉田,感谢重座本次航班。我们将典型到达,祝您路途愉快,谢谢。" }, { "language": "ja", "text": "こんにちは、私はチェチェンです。感謝の皆さんに来てくれました。私たちは、私たちの動画を楽しみに。ありがとうございました。" } ], "id": 124, "review_annotation": "待確認", "row_id": 124, "bounded_row_id": 124, "parent_row_ids": [ 124 ], "parent_row_id": 124, "packet_id": 47, "LID_evidence": [ { "region_id": 114, "probabilities": { "ja": 0.048260140465572476, "zh": 0.7744236052036285, "en": 0.03968652617186308 }, "status": "STABLE" }, { "region_id": 115, "probabilities": { "ja": 0.059596839104779066, "zh": 0.73847766071558, "en": 0.043367015477269885 }, "status": "STABLE" } ], "VAD_region_ids": [ 114, 115 ], "language_confidence": 0.7744236052036285, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 794.1, "packet_end": 807.5999999999999, "raw_text": "你好,我是吉田,感谢重座本次航班。", "selected_text": "你好，我是志全。感谢乘坐本次航班", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 124 ], "packet_id": 47, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "你好,我是吉田,感谢重座本次航班。", "after": "你好，我是志全。感谢乘坐本次航班", "classification": "BEST_SUPPORTED_WITH_REVIEW", "reason": "Independent ASR identifies name '志全' and verb '乘坐' (ride), correcting source '吉田' and '重座'." } ], "language": "zh", "packet_ids": [ 47 ], "index": 124, "independent_ASR": { "text": "你好，我是志全。感谢乘坐本次航班，我们乘坐的是到达，祝您旅途愉快，谢谢。", "chunks": [ { "text": "你好，我是志全。感谢乘坐本次航班，我们乘坐的是到达，祝您旅途愉快，谢谢。", "start": 0.0, "end": 13.5, "chunk_index": 0, "language": "Chinese", "finish_reason": "eos", "truncated": false, "generated_tokens": 23, "max_new_tokens": 194 } ], "detected_language": "Chinese", "segments": [ { "start": 794.1, "end": 807.6, "text": "你好，我是志全。感谢乘坐本次航班，我们乘坐的是到达，祝您旅途愉快，谢谢。" } ], "start": 794.1, "end": 807.5999999999999, "engine": "mlx-qwen3-asr", "language": "zh" }, "raw_word_timestamps": [ { "word": "你好,", "start": 794.76, "end": 795.4, "probability": 0.779296875 }, { "word": "我是", "start": 795.9200000000001, "end": 796.26, "probability": 0.95947265625 }, { "word": "吉", "start": 796.26, "end": 796.52, "probability": 0.106689453125 }, { "word": "田,", "start": 796.52, "end": 796.78, "probability": 0.11944580078125 }, { "word": "感", "start": 798.0, "end": 798.1, "probability": 0.4326171875 }, { "word": "谢", "start": 798.1, "end": 798.46, "probability": 0.994140625 }, { "word": "重", "start": 798.46, "end": 798.6800000000001, "probability": 0.145751953125 }, { "word": "座", "start": 798.6800000000001, "end": 799.0400000000001, "probability": 0.71826171875 }, { "word": "本", "start": 799.0400000000001, "end": 799.5, "probability": 0.65625 }, { "word": "次", "start": 799.5, "end": 799.8000000000001, "probability": 0.92724609375 }, { "word": "航", "start": 799.8000000000001, "end": 800.0, "probability": 0.8994140625 }, { "word": "班。", "start": 800.0, "end": 800.28, "probability": 0.99072265625 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "你好，我是志全。感谢乘坐本次航班", "review_status": "BEST_SUPPORTED_WITH_REVIEW", "classification": "BEST_SUPPORTED_WITH_REVIEW", "confidence": 0.85, "correction_reason": "Independent ASR identifies name '志全' and verb '乘坐' (ride), correcting source '吉田' and '重座'." }, { "start": 801.14, "end": 806.4, "text": "我们乘坐的是到达，祝您旅途愉快，谢谢。", "raw_asr_text": "我们将典型到达,祝您路途愉快,谢谢。", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 47, "words": [], "asr_metrics": { "avg_logprob": -0.5655769525572311, "compression_ratio": 0.9326923076923077, "no_speech_prob": 1.587830561478043e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "zh", "text": "你好,我是吉田,感谢重座本次航班。我们将典型到达,祝您路途愉快,谢谢。" }, { "language": "ja", "text": "こんにちは、私はチェチェンです。感謝の皆さんに来てくれました。私たちは、私たちの動画を楽しみに。ありがとうございました。" } ], "id": 125, "review_annotation": "待確認", "row_id": 125, "bounded_row_id": 125, "parent_row_ids": [ 125 ], "parent_row_id": 125, "packet_id": 47, "LID_evidence": [ { "region_id": 116, "probabilities": { "ja": 0.2544611095217988, "zh": 0.15094979479908943, "en": 0.0862426208332181 }, "status": "UNCERTAIN" }, { "region_id": 117, "probabilities": { "ja": 0.5224312717715899, "zh": 0.12582204894473156, "en": 0.05942585660765568 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 116, 117 ], "language_confidence": 0.15094979479908943, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 794.1, "packet_end": 807.5999999999999, "raw_text": "我们将典型到达,祝您路途愉快,谢谢。", "selected_text": "我们乘坐的是到达，祝您旅途愉快，谢谢。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 125 ], "packet_id": 47, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "我们将典型到达,祝您路途愉快,谢谢。", "after": "我们乘坐的是到达，祝您旅途愉快，谢谢。", "classification": "BEST_SUPPORTED_WITH_REVIEW", "reason": "Independent ASR provides coherent phrase '乘坐的是到达' (likely 'soon arriving'), correcting source nonsense." } ], "language": "zh", "packet_ids": [ 47 ], "index": 125, "independent_ASR": { "text": "你好，我是志全。感谢乘坐本次航班，我们乘坐的是到达，祝您旅途愉快，谢谢。", "chunks": [ { "text": "你好，我是志全。感谢乘坐本次航班，我们乘坐的是到达，祝您旅途愉快，谢谢。", "start": 0.0, "end": 13.5, "chunk_index": 0, "language": "Chinese", "finish_reason": "eos", "truncated": false, "generated_tokens": 23, "max_new_tokens": 194 } ], "detected_language": "Chinese", "segments": [ { "start": 794.1, "end": 807.6, "text": "你好，我是志全。感谢乘坐本次航班，我们乘坐的是到达，祝您旅途愉快，谢谢。" } ], "start": 794.1, "end": 807.5999999999999, "engine": "mlx-qwen3-asr", "language": "zh" }, "raw_word_timestamps": [ { "word": "我们", "start": 801.14, "end": 801.78, "probability": 0.97509765625 }, { "word": "将", "start": 801.78, "end": 802.22, "probability": 0.607421875 }, { "word": "典", "start": 802.22, "end": 802.5400000000001, "probability": 0.5732421875 }, { "word": "型", "start": 802.5400000000001, "end": 802.88, "probability": 0.625 }, { "word": "到", "start": 802.88, "end": 803.2, "probability": 0.462890625 }, { "word": "达,", "start": 803.2, "end": 803.52, "probability": 0.78076171875 }, { "word": "祝", "start": 804.3000000000001, "end": 804.5, "probability": 0.986328125 }, { "word": "您", "start": 804.5, "end": 804.72, "probability": 0.96875 }, { "word": "路", "start": 804.72, "end": 805.0400000000001, "probability": 0.390380859375 }, { "word": "途", "start": 805.0400000000001, "end": 805.38, "probability": 0.921875 }, { "word": "愉", "start": 805.38, "end": 805.76, "probability": 0.9443359375 }, { "word": "快,", "start": 805.76, "end": 806.02, "probability": 0.99951171875 }, { "word": "谢谢。", "start": 806.2, "end": 806.4, "probability": 0.98583984375 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "我们乘坐的是到达，祝您旅途愉快，谢谢。", "review_status": "BEST_SUPPORTED_WITH_REVIEW", "classification": "BEST_SUPPORTED_WITH_REVIEW", "confidence": 0.85, "correction_reason": "Independent ASR provides coherent phrase '乘坐的是到达' (likely 'soon arriving'), correcting source nonsense." }, { "start": 811.9, "end": 814.4, "text": "着陸まではおよそ1時間ほどの飛行でございます。", "raw_asr_text": "着陸まではおよそ1時間ほどの飛行でございます。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 48, "words": [], "asr_metrics": { "avg_logprob": -0.18471784591674806, "compression_ratio": 0.9178082191780822, "no_speech_prob": 6.465470520078398e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "着陸まではおよそ1時間ほどの飛行でございます。" } ], "id": 126, "review_annotation": "", "row_id": 126, "bounded_row_id": 126, "parent_row_ids": [ 126 ], "parent_row_id": 126, "packet_id": 48, "LID_evidence": [ { "region_id": 118, "probabilities": { "ja": 0.9833823442459106, "zh": 0.0011389878345653415, "en": 0.009783314618592462 }, "status": "STABLE" } ], "VAD_region_ids": [ 118 ], "language_confidence": 0.9833823442459106, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 811.1, "packet_end": 815.6999999999999, "raw_text": "着陸まではおよそ1時間ほどの飛行でございます。", "selected_text": "着陸まではおよそ1時間ほどの飛行でございます。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 126 ], "packet_id": 48, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "着陸まではおよそ1時間ほどの飛行でございます。", "after": "着陸まではおよそ1時間ほどの飛行でございます。", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Clear Japanese announcement matching standard phrasing." } ], "language": "ja", "packet_ids": [ 48 ], "index": 126, "source_text": "着陸まではおよそ1時間ほどの飛行でございます。", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Clear Japanese announcement matching standard phrasing.", "raw_word_timestamps": [ { "word": "着", "start": 811.1, "end": 811.4200000000001, "probability": 0.205810546875 }, { "word": "陸", "start": 811.4200000000001, "end": 811.76, "probability": 0.9990234375 }, { "word": "ま", "start": 811.76, "end": 811.9, "probability": 0.998046875 }, { "word": "では", "start": 811.9, "end": 812.16, "probability": 0.99951171875 }, { "word": "お", "start": 812.16, "end": 812.34, "probability": 0.9150390625 }, { "word": "よ", "start": 812.34, "end": 812.46, "probability": 1.0 }, { "word": "そ", "start": 812.46, "end": 812.6, "probability": 1.0 }, { "word": "1", "start": 812.6, "end": 812.8000000000001, "probability": 0.99462890625 }, { "word": "時間", "start": 812.8000000000001, "end": 813.1, "probability": 1.0 }, { "word": "ほど", "start": 813.1, "end": 813.34, "probability": 0.990234375 }, { "word": "の", "start": 813.34, "end": 813.5, "probability": 0.9990234375 }, { "word": "飛", "start": 813.5, "end": 813.6, "probability": 0.9951171875 }, { "word": "行", "start": 813.6, "end": 813.76, "probability": 0.9990234375 }, { "word": "で", "start": 813.76, "end": 813.88, "probability": 0.98291015625 }, { "word": "ございます。", "start": 813.88, "end": 814.4, "probability": 0.99951171875 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 817.9, "end": 820.2, "text": "化粧室はお早めにご使用ください", "raw_asr_text": "化粧室はお早めにご使用ください", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 49, "words": [], "asr_metrics": { "avg_logprob": -0.21812980315264532, "compression_ratio": 0.8035714285714286, "no_speech_prob": 4.464314334873265e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "化粧室はお早めにご使用ください" } ], "id": 127, "review_annotation": "", "row_id": 127, "bounded_row_id": 127, "parent_row_ids": [ 127 ], "parent_row_id": 127, "packet_id": 49, "LID_evidence": [ { "region_id": 119, "probabilities": { "ja": 0.9956286748250326, "zh": 0.00028471293626353145, "en": 0.0021197875418389836 }, "status": "STABLE" } ], "VAD_region_ids": [ 119 ], "language_confidence": 0.9956286748250326, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 817.1, "packet_end": 821.5999999999999, "raw_text": "化粧室はお早めにご使用ください", "selected_text": "化粧室はお早めにご使用ください", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 127 ], "packet_id": 49, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "化粧室はお早めにご使用ください", "after": "化粧室はお早めにご使用ください", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard Japanese announcement." } ], "language": "ja", "packet_ids": [ 49 ], "index": 127, "source_text": "化粧室はお早めにご使用ください", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Standard Japanese announcement.", "raw_word_timestamps": [ { "word": "化", "start": 817.84, "end": 818.2, "probability": 0.98583984375 }, { "word": "粧", "start": 818.2, "end": 818.38, "probability": 0.99951171875 }, { "word": "室", "start": 818.38, "end": 818.5600000000001, "probability": 0.98046875 }, { "word": "は", "start": 818.5600000000001, "end": 818.74, "probability": 0.99951171875 }, { "word": "お", "start": 818.74, "end": 818.9200000000001, "probability": 0.8955078125 }, { "word": "早", "start": 818.9200000000001, "end": 819.08, "probability": 0.99951171875 }, { "word": "め", "start": 819.08, "end": 819.24, "probability": 0.99951171875 }, { "word": "に", "start": 819.24, "end": 819.4200000000001, "probability": 1.0 }, { "word": "ご", "start": 819.4200000000001, "end": 819.58, "probability": 0.998046875 }, { "word": "使", "start": 819.58, "end": 819.7, "probability": 0.99853515625 }, { "word": "用", "start": 819.7, "end": 819.84, "probability": 0.99951171875 }, { "word": "ください", "start": 819.84, "end": 820.2, "probability": 0.99951171875 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 825.8, "end": 829.08, "text": "日本入国書類のご記入はお済みでしょうか", "raw_asr_text": "日本入国書類のご記入はお済みでしょうか", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 50, "words": [], "asr_metrics": { "avg_logprob": -0.16729845539216073, "compression_ratio": 1.0736842105263158, "no_speech_prob": 5.258811181657608e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "日本入国書類のご記入はお済みでしょうかどうぞもう一度お確かめください" } ], "id": 128, "review_annotation": "", "row_id": 128, "bounded_row_id": 128, "parent_row_ids": [ 128 ], "parent_row_id": 128, "packet_id": 50, "LID_evidence": [ { "region_id": 120, "probabilities": { "ja": 0.9878771066665649, "zh": 0.000818207353586331, "en": 0.006244580983184278 }, "status": "STABLE" } ], "VAD_region_ids": [ 120 ], "language_confidence": 0.9878771066665649, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 825.0, "packet_end": 832.9, "raw_text": "日本入国書類のご記入はお済みでしょうか", "selected_text": "日本入国書類のご記入はお済みでしょうか", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 128 ], "packet_id": 50, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "日本入国書類のご記入はお済みでしょうか", "after": "日本入国書類のご記入はお済みでしょうか", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard Japanese announcement." } ], "language": "ja", "packet_ids": [ 50 ], "index": 128, "source_text": "日本入国書類のご記入はお済みでしょうか", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Standard Japanese announcement.", "raw_word_timestamps": [ { "word": "日本", "start": 825.76, "end": 826.16, "probability": 0.97998046875 }, { "word": "入", "start": 826.16, "end": 826.5, "probability": 0.998046875 }, { "word": "国", "start": 826.5, "end": 826.78, "probability": 0.99951171875 }, { "word": "書", "start": 826.78, "end": 827.0, "probability": 0.9990234375 }, { "word": "類", "start": 827.0, "end": 827.22, "probability": 0.99951171875 }, { "word": "の", "start": 827.22, "end": 827.42, "probability": 1.0 }, { "word": "ご", "start": 827.42, "end": 827.64, "probability": 0.99853515625 }, { "word": "記", "start": 827.64, "end": 827.74, "probability": 0.9990234375 }, { "word": "入", "start": 827.74, "end": 827.94, "probability": 1.0 }, { "word": "は", "start": 827.94, "end": 828.12, "probability": 1.0 }, { "word": "お", "start": 828.12, "end": 828.3, "probability": 0.99609375 }, { "word": "済", "start": 828.3, "end": 828.46, "probability": 0.9990234375 }, { "word": "み", "start": 828.46, "end": 828.54, "probability": 0.9990234375 }, { "word": "でしょう", "start": 828.54, "end": 828.8, "probability": 0.9990234375 }, { "word": "か", "start": 828.8, "end": 829.08, "probability": 1.0 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 829.08, "end": 831.5, "text": "どうぞもう一度お確かめください", "raw_asr_text": "どうぞもう一度お確かめください", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 50, "words": [ { "word": "どう", "start": 829.08, "end": 829.82, "probability": 0.1114501953125 }, { "word": "ぞ", "start": 829.82, "end": 829.98, "probability": 1.0 }, { "word": "もう", "start": 829.98, "end": 830.14, "probability": 0.998046875 }, { "word": "一", "start": 830.14, "end": 830.34, "probability": 1.0 }, { "word": "度", "start": 830.34, "end": 830.52, "probability": 1.0 }, { "word": "お", "start": 830.52, "end": 830.72, "probability": 0.9970703125 }, { "word": "確", "start": 830.72, "end": 830.84, "probability": 0.9990234375 }, { "word": "か", "start": 830.84, "end": 831.0, "probability": 0.99951171875 }, { "word": "め", "start": 831.0, "end": 831.1, "probability": 0.99560546875 }, { "word": "ください", "start": 831.1, "end": 831.5, "probability": 0.99853515625 } ], "asr_metrics": { "avg_logprob": -0.16729845539216073, "compression_ratio": 1.0736842105263158, "no_speech_prob": 5.258811181657608e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "日本入国書類のご記入はお済みでしょうかどうぞもう一度お確かめください" } ], "id": 129, "review_annotation": "", "row_id": 129, "bounded_row_id": 129, "parent_row_ids": [ 129 ], "parent_row_id": 129, "packet_id": 50, "LID_evidence": [ { "region_id": 120, "probabilities": { "ja": 0.9878771066665649, "zh": 0.000818207353586331, "en": 0.006244580983184278 }, "status": "STABLE" } ], "VAD_region_ids": [ 120 ], "language_confidence": 0.9878771066665649, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 825.0, "packet_end": 832.9, "raw_text": "どうぞもう一度お確かめください", "selected_text": "どうぞもう一度お確かめください", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 129 ], "packet_id": 50, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "どうぞもう一度お確かめください", "after": "どうぞもう一度お確かめください", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard Japanese announcement." } ], "language": "ja", "packet_ids": [ 50 ], "index": 129, "source_text": "どうぞもう一度お確かめください", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Standard Japanese announcement." }, { "start": 845.3, "end": 847.72, "text": "到着地の天候は雨、気温は7度でございます。", "raw_asr_text": "到着地の天候は雨、気温は7度でございます。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 51, "words": [], "asr_metrics": { "avg_logprob": -0.07077116057986305, "compression_ratio": 1.076271186440678, "no_speech_prob": 4.921753109232707e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "到着地の天候は雨、気温は7度でございます。国際線ターミナル111番ゲートに到着いたします。" } ], "id": 130, "review_annotation": "", "row_id": 130, "bounded_row_id": 130, "parent_row_ids": [ 130 ], "parent_row_id": 130, "packet_id": 51, "LID_evidence": [ { "region_id": 121, "probabilities": { "ja": 0.9466931720574697, "zh": 0.004364678097772412, "en": 0.02373223553877324 }, "status": "STABLE" } ], "VAD_region_ids": [ 121 ], "language_confidence": 0.9466931720574697, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 844.5, "packet_end": 853.5, "raw_text": "到着地の天候は雨、気温は7度でございます。", "selected_text": "到着地の天候は雨、気温は7度でございます。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 130 ], "packet_id": 51, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "到着地の天候は雨、気温は7度でございます。", "after": "到着地の天候は雨、気温は7度でございます。", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Text matches standard Japanese announcement phrasing and context perfectly." } ], "language": "ja", "packet_ids": [ 51 ], "index": 130, "source_text": "到着地の天候は雨、気温は7度でございます。", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Text matches standard Japanese announcement phrasing and context perfectly.", "raw_word_timestamps": [ { "word": "到", "start": 844.5, "end": 844.86, "probability": 0.9931640625 }, { "word": "着", "start": 844.86, "end": 845.04, "probability": 1.0 }, { "word": "地", "start": 845.04, "end": 845.18, "probability": 0.98291015625 }, { "word": "の", "start": 845.18, "end": 845.32, "probability": 1.0 }, { "word": "天", "start": 845.32, "end": 845.5, "probability": 0.99951171875 }, { "word": "候", "start": 845.5, "end": 845.74, "probability": 0.9970703125 }, { "word": "は", "start": 845.74, "end": 845.92, "probability": 1.0 }, { "word": "雨、", "start": 845.92, "end": 846.2, "probability": 0.98779296875 }, { "word": "気", "start": 846.36, "end": 846.46, "probability": 0.99267578125 }, { "word": "温", "start": 846.46, "end": 846.58, "probability": 1.0 }, { "word": "は", "start": 846.58, "end": 846.7, "probability": 0.9990234375 }, { "word": "7", "start": 846.7, "end": 846.86, "probability": 0.998046875 }, { "word": "度", "start": 846.86, "end": 847.04, "probability": 0.99951171875 }, { "word": "で", "start": 847.04, "end": 847.16, "probability": 0.9990234375 }, { "word": "ございます。", "start": 847.16, "end": 847.72, "probability": 1.0 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 848.76, "end": 852.26, "text": "国際線ターミナル111番ゲートに到着いたします。", "raw_asr_text": "国際線ターミナル111番ゲートに到着いたします。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 51, "words": [ { "word": "国", "start": 848.76, "end": 849.12, "probability": 0.9921875 }, { "word": "際", "start": 849.12, "end": 849.24, "probability": 0.99951171875 }, { "word": "線", "start": 849.24, "end": 849.48, "probability": 0.99853515625 }, { "word": "タ", "start": 849.48, "end": 849.64, "probability": 0.998046875 }, { "word": "ーミ", "start": 849.64, "end": 849.78, "probability": 0.9990234375 }, { "word": "ナ", "start": 849.78, "end": 849.9, "probability": 0.99951171875 }, { "word": "ル", "start": 849.9, "end": 850.02, "probability": 1.0 }, { "word": "11", "start": 850.02, "end": 850.36, "probability": 0.96044921875 }, { "word": "1", "start": 850.36, "end": 850.6, "probability": 0.99951171875 }, { "word": "番", "start": 850.6, "end": 850.82, "probability": 0.99853515625 }, { "word": "ゲ", "start": 850.82, "end": 850.98, "probability": 0.99951171875 }, { "word": "ート", "start": 850.98, "end": 851.12, "probability": 1.0 }, { "word": "に", "start": 851.12, "end": 851.26, "probability": 1.0 }, { "word": "到", "start": 851.26, "end": 851.46, "probability": 0.99951171875 }, { "word": "着", "start": 851.46, "end": 851.7, "probability": 1.0 }, { "word": "いた", "start": 851.7, "end": 851.88, "probability": 0.998046875 }, { "word": "します。", "start": 851.88, "end": 852.26, "probability": 1.0 } ], "asr_metrics": { "avg_logprob": -0.07077116057986305, "compression_ratio": 1.076271186440678, "no_speech_prob": 4.921753109232707e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "到着地の天候は雨、気温は7度でございます。国際線ターミナル111番ゲートに到着いたします。" } ], "id": 131, "review_annotation": "", "row_id": 131, "bounded_row_id": 131, "parent_row_ids": [ 131 ], "parent_row_id": 131, "packet_id": 51, "LID_evidence": [ { "region_id": 122, "probabilities": { "ja": 0.9764988049864769, "zh": 0.0017938918699655915, "en": 0.012747262018820038 }, "status": "STABLE" } ], "VAD_region_ids": [ 122 ], "language_confidence": 0.9764988049864769, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 844.5, "packet_end": 853.5, "raw_text": "国際線ターミナル111番ゲートに到着いたします。", "selected_text": "国際線ターミナル111番ゲートに到着いたします。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 131 ], "packet_id": 51, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "国際線ターミナル111番ゲートに到着いたします。", "after": "国際線ターミナル111番ゲートに到着いたします。", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Text matches standard Japanese announcement phrasing and context perfectly." } ], "language": "ja", "packet_ids": [ 51 ], "index": 131, "source_text": "国際線ターミナル111番ゲートに到着いたします。", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Text matches standard Japanese announcement phrasing and context perfectly." }, { "start": 856.1, "end": 858.9200000000001, "text": "Ladies and gentlemen, we'll be landing in about one hour.", "raw_asr_text": "Ladies and gentlemen, we'll be landing in about one hour.", "source_language": "en", "source_lang": "en", "asr_packet_id": 52, "words": [], "asr_metrics": { "avg_logprob": -0.1831723873431866, "compression_ratio": 1.4134078212290502, "no_speech_prob": 3.76983513850071e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "en", "text": "Ladies and gentlemen, we'll be landing in about one hour. Passengers who wish to use the laboratory are kindly asked to do so at this time. We'll be arriving at the gate number 111. The weather is raining and temperature is 7 degrees Celsius. Thank you." }, { "language": "ja", "text": "Ladies and gentlemen, we'll be landing in about 1 hour.Passengers who wish to use the laboratory are kindly asked to do so at this time.We'll be arriving at the gate number 111.The weather is raining and temperature is 7 degrees Celsius. Thank you." } ], "id": 132, "review_annotation": "", "row_id": 132, "bounded_row_id": 132, "parent_row_ids": [ 132 ], "parent_row_id": 132, "packet_id": 52, "LID_evidence": [ { "region_id": 123, "probabilities": { "ja": 0.020750315760129263, "zh": 0.004700702781389866, "en": 0.754328898021153 }, "status": "STABLE" } ], "VAD_region_ids": [ 123 ], "language_confidence": 0.754328898021153, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 855.3000000000001, "packet_end": 878.8, "raw_text": "Ladies and gentlemen, we'll be landing in about one hour.", "selected_text": "Ladies and gentlemen, we'll be landing in about one hour.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 132 ], "packet_id": 52, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "Ladies and gentlemen, we'll be landing in about one hour.", "after": "Ladies and gentlemen, we'll be landing in about one hour.", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Text matches standard English announcement phrasing and context perfectly." } ], "language": "en", "packet_ids": [ 52 ], "index": 132, "source_text": "Ladies and gentlemen, we'll be landing in about one hour.", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Text matches standard English announcement phrasing and context perfectly.", "raw_word_timestamps": [ { "word": " Ladies", "start": 855.3000000000001, "end": 856.46, "probability": 0.84375 }, { "word": " and", "start": 856.46, "end": 856.7400000000001, "probability": 0.98974609375 }, { "word": " gentlemen,", "start": 856.7400000000001, "end": 857.0600000000001, "probability": 0.61572265625 }, { "word": " we'll", "start": 857.32, "end": 857.44, "probability": 0.83203125 }, { "word": " be", "start": 857.44, "end": 857.5200000000001, "probability": 0.998046875 }, { "word": " landing", "start": 857.5200000000001, "end": 857.86, "probability": 0.9609375 }, { "word": " in", "start": 857.86, "end": 858.1, "probability": 0.52001953125 }, { "word": " about", "start": 858.1, "end": 858.2800000000001, "probability": 0.99365234375 }, { "word": " one", "start": 858.2800000000001, "end": 858.58, "probability": 0.591796875 }, { "word": " hour.", "start": 858.58, "end": 858.9200000000001, "probability": 0.998046875 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 859.48, "end": 864.5000000000001, "text": "Passengers who wish to use the lavatory are kindly asked to do so at this time.", "raw_asr_text": "Passengers who wish to use the laboratory are kindly asked to do so at this time.", "source_language": "en", "source_lang": "en", "asr_packet_id": 52, "words": [], "asr_metrics": { "avg_logprob": -0.1831723873431866, "compression_ratio": 1.4134078212290502, "no_speech_prob": 3.76983513850071e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "en", "text": "Ladies and gentlemen, we'll be landing in about one hour. Passengers who wish to use the laboratory are kindly asked to do so at this time. We'll be arriving at the gate number 111. The weather is raining and temperature is 7 degrees Celsius. Thank you." }, { "language": "ja", "text": "Ladies and gentlemen, we'll be landing in about 1 hour.Passengers who wish to use the laboratory are kindly asked to do so at this time.We'll be arriving at the gate number 111.The weather is raining and temperature is 7 degrees Celsius. Thank you." } ], "id": 133, "review_annotation": "", "row_id": 133, "bounded_row_id": 133, "parent_row_ids": [ 133 ], "parent_row_id": 133, "packet_id": 52, "LID_evidence": [ { "region_id": 124, "probabilities": { "ja": 0.03352880373131484, "zh": 0.003834973875200376, "en": 0.8247351348400116 }, "status": "STABLE" } ], "VAD_region_ids": [ 124 ], "language_confidence": 0.8247351348400116, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 855.3000000000001, "packet_end": 878.8, "raw_text": "Passengers who wish to use the laboratory are kindly asked to do so at this time.", "selected_text": "Passengers who wish to use the lavatory are kindly asked to do so at this time.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 133 ], "packet_id": 52, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "Passengers who wish to use the laboratory are kindly asked to do so at this time.", "after": "Passengers who wish to use the lavatory are kindly asked to do so at this time.", "classification": "CERTAIN_RESTORATION", "reason": "Independent ASR confirms 'lavatory' and context supports standard safety announcement phrasing over 'laboratory'." } ], "language": "en", "packet_ids": [ 52 ], "index": 133, "independent_ASR": { "text": "Ladies and gentlemen, we'll be landing about one hour. Passengers who wish to use the lavatory are kindly asked to do so at this time. We'll be arriving at the gate number 111. The weather is raining and temperature is seven degrees Celsius. Thank you.", "chunks": [ { "text": "Ladies and gentlemen, we'll be landing about one hour. Passengers who wish to use the lavatory are kindly asked to do so at this time. We'll be arriving at the gate number 111. The weather is raining and temperature is seven degrees Celsius. Thank you.", "start": 0.0, "end": 23.5, "chunk_index": 0, "language": "English", "finish_reason": "eos", "truncated": false, "generated_tokens": 59, "max_new_tokens": 314 } ], "detected_language": "English", "segments": [ { "start": 855.3000000000001, "end": 878.8000000000001, "text": "Ladies and gentlemen, we'll be landing about one hour. Passengers who wish to use the lavatory are kindly asked to do so at this time. We'll be arriving at the gate number 111. The weather is raining and temperature is seven degrees Celsius. Thank you." } ], "start": 855.3000000000001, "end": 878.8, "engine": "mlx-qwen3-asr", "language": "en" }, "raw_word_timestamps": [ { "word": " Passengers", "start": 859.48, "end": 860.1400000000001, "probability": 0.978515625 }, { "word": " who", "start": 860.1400000000001, "end": 860.5000000000001, "probability": 0.96337890625 }, { "word": " wish", "start": 860.5000000000001, "end": 860.7800000000001, "probability": 0.99267578125 }, { "word": " to", "start": 860.7800000000001, "end": 860.98, "probability": 0.9990234375 }, { "word": " use", "start": 860.98, "end": 861.32, "probability": 0.9951171875 }, { "word": " the", "start": 861.32, "end": 861.5000000000001, "probability": 0.72705078125 }, { "word": " laboratory", "start": 861.5000000000001, "end": 861.9200000000001, "probability": 0.646484375 }, { "word": " are", "start": 861.9200000000001, "end": 862.36, "probability": 0.9794921875 }, { "word": " kindly", "start": 862.36, "end": 862.6, "probability": 0.99755859375 }, { "word": " asked", "start": 862.6, "end": 863.0600000000001, "probability": 0.97216796875 }, { "word": " to", "start": 863.0600000000001, "end": 863.22, "probability": 0.98974609375 }, { "word": " do", "start": 863.22, "end": 863.36, "probability": 0.994140625 }, { "word": " so", "start": 863.36, "end": 863.6200000000001, "probability": 0.99853515625 }, { "word": " at", "start": 863.6200000000001, "end": 863.8800000000001, "probability": 0.99560546875 }, { "word": " this", "start": 863.8800000000001, "end": 864.1400000000001, "probability": 0.9990234375 }, { "word": " time.", "start": 864.1400000000001, "end": 864.5000000000001, "probability": 0.99951171875 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "Passengers who wish to use the lavatory are kindly asked to do so at this time.", "review_status": "CERTAIN_RESTORATION", "classification": "CERTAIN_RESTORATION", "confidence": 0.95, "correction_reason": "Independent ASR confirms 'lavatory' and context supports standard safety announcement phrasing over 'laboratory'." }, { "start": 866.08, "end": 872.0600000000001, "text": "We'll be arriving at the gate number 111.", "raw_asr_text": "We'll be arriving at the gate number 111.", "source_language": "en", "source_lang": "en", "asr_packet_id": 52, "words": [ { "word": " We'll", "start": 866.08, "end": 866.7400000000001, "probability": 0.9619140625 }, { "word": " be", "start": 866.7400000000001, "end": 866.84, "probability": 0.998046875 }, { "word": " arriving", "start": 866.84, "end": 867.22, "probability": 0.9990234375 }, { "word": " at", "start": 867.22, "end": 867.5200000000001, "probability": 0.99755859375 }, { "word": " the", "start": 867.5200000000001, "end": 867.6200000000001, "probability": 0.83740234375 }, { "word": " gate", "start": 867.6200000000001, "end": 867.86, "probability": 0.98193359375 }, { "word": " number", "start": 867.86, "end": 868.2800000000001, "probability": 0.9755859375 }, { "word": " 111.", "start": 868.2800000000001, "end": 872.0600000000001, "probability": 0.8955078125 } ], "asr_metrics": { "avg_logprob": -0.1831723873431866, "compression_ratio": 1.4134078212290502, "no_speech_prob": 3.76983513850071e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "en", "text": "Ladies and gentlemen, we'll be landing in about one hour. Passengers who wish to use the laboratory are kindly asked to do so at this time. We'll be arriving at the gate number 111. The weather is raining and temperature is 7 degrees Celsius. Thank you." }, { "language": "ja", "text": "Ladies and gentlemen, we'll be landing in about 1 hour.Passengers who wish to use the laboratory are kindly asked to do so at this time.We'll be arriving at the gate number 111.The weather is raining and temperature is 7 degrees Celsius. Thank you." } ], "id": 134, "review_annotation": "", "row_id": 134, "bounded_row_id": 134, "parent_row_ids": [ 134 ], "parent_row_id": 134, "packet_id": 52, "LID_evidence": [ { "region_id": 125, "probabilities": { "ja": 0.00435369077604264, "zh": 0.002583804482128471, "en": 0.9402436316013336 }, "status": "UNCERTAIN" }, { "region_id": 126, "probabilities": { "ja": 0.029462510021403432, "zh": 0.009463754249736667, "en": 0.8022177070379257 }, "status": "STABLE" } ], "VAD_region_ids": [ 125, 126 ], "language_confidence": 0.9402436316013336, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 855.3000000000001, "packet_end": 878.8, "raw_text": "We'll be arriving at the gate number 111.", "selected_text": "We'll be arriving at the gate number 111.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 134 ], "packet_id": 52, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "We'll be arriving at the gate number 111.", "after": "We'll be arriving at the gate number 111.", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Text matches standard English announcement phrasing and context perfectly." } ], "language": "en", "packet_ids": [ 52 ], "index": 134, "source_text": "We'll be arriving at the gate number 111.", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Text matches standard English announcement phrasing and context perfectly." }, { "start": 872.6400000000001, "end": 877.5400000000001, "text": "The weather is raining and temperature is 7 degrees Celsius. Thank you.", "raw_asr_text": "The weather is raining and temperature is 7 degrees Celsius. Thank you.", "source_language": "en", "source_lang": "en", "asr_packet_id": 52, "words": [ { "word": " The", "start": 872.6400000000001, "end": 873.3000000000001, "probability": 0.99072265625 }, { "word": " weather", "start": 873.3000000000001, "end": 873.5400000000001, "probability": 0.998046875 }, { "word": " is", "start": 873.5400000000001, "end": 873.86, "probability": 0.9990234375 }, { "word": " raining", "start": 873.86, "end": 874.2400000000001, "probability": 0.9833984375 }, { "word": " and", "start": 874.2400000000001, "end": 874.5000000000001, "probability": 0.931640625 }, { "word": " temperature", "start": 874.5000000000001, "end": 874.9000000000001, "probability": 0.96484375 }, { "word": " is", "start": 874.9000000000001, "end": 875.3000000000001, "probability": 0.99755859375 }, { "word": " 7", "start": 875.3000000000001, "end": 875.6600000000001, "probability": 0.828125 }, { "word": " degrees", "start": 875.6600000000001, "end": 876.0600000000001, "probability": 0.861328125 }, { "word": " Celsius.", "start": 876.0600000000001, "end": 876.58, "probability": 0.9423828125 }, { "word": " Thank", "start": 876.98, "end": 877.3000000000001, "probability": 0.99609375 }, { "word": " you.", "start": 877.3000000000001, "end": 877.5400000000001, "probability": 0.98828125 } ], "asr_metrics": { "avg_logprob": -0.1831723873431866, "compression_ratio": 1.4134078212290502, "no_speech_prob": 3.76983513850071e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "en", "text": "Ladies and gentlemen, we'll be landing in about one hour. Passengers who wish to use the laboratory are kindly asked to do so at this time. We'll be arriving at the gate number 111. The weather is raining and temperature is 7 degrees Celsius. Thank you." }, { "language": "ja", "text": "Ladies and gentlemen, we'll be landing in about 1 hour.Passengers who wish to use the laboratory are kindly asked to do so at this time.We'll be arriving at the gate number 111.The weather is raining and temperature is 7 degrees Celsius. Thank you." } ], "id": 135, "review_annotation": "", "row_id": 135, "bounded_row_id": 135, "parent_row_ids": [ 135 ], "parent_row_id": 135, "packet_id": 52, "LID_evidence": [ { "region_id": 127, "probabilities": { "ja": 0.0290865630377084, "zh": 0.00686712961178273, "en": 0.872191971540451 }, "status": "STABLE" } ], "VAD_region_ids": [ 127 ], "language_confidence": 0.872191971540451, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 855.3000000000001, "packet_end": 878.8, "raw_text": "The weather is raining and temperature is 7 degrees Celsius. Thank you.", "selected_text": "The weather is raining and temperature is 7 degrees Celsius. Thank you.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 135 ], "packet_id": 52, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "The weather is raining and temperature is 7 degrees Celsius. Thank you.", "after": "The weather is raining and temperature is 7 degrees Celsius. Thank you.", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Text matches standard English announcement phrasing and context perfectly." } ], "language": "en", "packet_ids": [ 52 ], "index": 135, "source_text": "The weather is raining and temperature is 7 degrees Celsius. Thank you.", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Text matches standard English announcement phrasing and context perfectly." }, { "start": 881.7, "end": 888.6400000000001, "text": "各位贵宾,飞机预计将在一个小时后抵达东京羽田国际机场", "raw_asr_text": "各位贵宾,飞机预计将在一个小时后抵达东京雨田国际机场", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 53, "words": [], "asr_metrics": { "avg_logprob": -0.11775940981778231, "compression_ratio": 1.172093023255814, "no_speech_prob": 1.9891798277993367e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "zh", "text": "各位贵宾,飞机预计将在一个小时后抵达东京雨田国际机场当地的天气为雨天,气温为7度请您再次确认日本的入境资料是否已经填写好持有它国护照的旅客必须要填写入国登记卡以及海关申报单" }, { "language": "ja", "text": "ご視聴ありがとうございました。" } ], "id": 136, "review_annotation": "", "row_id": 136, "bounded_row_id": 136, "parent_row_ids": [ 136 ], "parent_row_id": 136, "packet_id": 53, "LID_evidence": [ { "region_id": 128, "probabilities": { "ja": 0.034557516453787684, "zh": 0.3882526258627574, "en": 0.15701413868616024 }, "status": "UNCERTAIN" }, { "region_id": 129, "probabilities": { "ja": 0.011427161190658808, "zh": 0.8309851984182993, "en": 0.06099555113663276 }, "status": "STABLE" } ], "VAD_region_ids": [ 128, 129 ], "language_confidence": 0.8309851984182993, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 880.9000000000001, "packet_end": 906.9, "raw_text": "各位贵宾,飞机预计将在一个小时后抵达东京雨田国际机场", "selected_text": "各位贵宾,飞机预计将在一个小时后抵达东京羽田国际机场", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 136 ], "packet_id": 53, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "各位贵宾,飞机预计将在一个小时后抵达东京雨田国际机场", "after": "各位贵宾,飞机预计将在一个小时后抵达东京羽田国际机场", "classification": "CERTAIN_RESTORATION", "reason": "Independent ASR confirms '羽田' (Haneda) correcting the ASR error '雨田' (Uta), consistent with flight destination." } ], "language": "zh", "packet_ids": [ 53 ], "index": 136, "independent_ASR": { "text": "各位贵宾，飞机预计将在一个小时后抵达东京羽田国际机场。当地的天气为雨天，气温为七度。请您再次确认日本的入境资料是否已经填写好。持有他国护照的旅客必须要填写，如果登机卡以及海关申报单。", "chunks": [ { "text": "各位贵宾，飞机预计将在一个小时后抵达东京羽田国际机场。当地的天气为雨天，气温为七度。请您再次确认日本的入境资料是否已经填写好。持有他国护照的旅客必须要填写，如果登机卡以及海关申报单。", "start": 0.0, "end": 26.0, "chunk_index": 0, "language": "Chinese", "finish_reason": "eos", "truncated": false, "generated_tokens": 55, "max_new_tokens": 344 } ], "detected_language": "Chinese", "segments": [ { "start": 880.9000000000001, "end": 906.9000000000001, "text": "各位贵宾，飞机预计将在一个小时后抵达东京羽田国际机场。当地的天气为雨天，气温为七度。请您再次确认日本的入境资料是否已经填写好。持有他国护照的旅客必须要填写，如果登机卡以及海关申报单。" } ], "start": 880.9000000000001, "end": 906.9, "engine": "mlx-qwen3-asr", "language": "zh" }, "raw_word_timestamps": [ { "word": "各位", "start": 881.5400000000001, "end": 881.9800000000001, "probability": 0.97509765625 }, { "word": "贵", "start": 881.9800000000001, "end": 882.2200000000001, "probability": 0.890625 }, { "word": "宾,", "start": 882.2200000000001, "end": 882.4200000000001, "probability": 0.97900390625 }, { "word": "飞", "start": 882.9000000000001, "end": 883.32, "probability": 0.9814453125 }, { "word": "机", "start": 883.32, "end": 883.5000000000001, "probability": 0.99853515625 }, { "word": "预", "start": 883.5000000000001, "end": 883.7800000000001, "probability": 0.9609375 }, { "word": "计", "start": 883.7800000000001, "end": 883.94, "probability": 0.998046875 }, { "word": "将", "start": 883.94, "end": 884.1400000000001, "probability": 0.99609375 }, { "word": "在", "start": 884.1400000000001, "end": 884.3800000000001, "probability": 0.99560546875 }, { "word": "一个", "start": 884.3800000000001, "end": 884.7, "probability": 0.9794921875 }, { "word": "小", "start": 884.7, "end": 884.96, "probability": 0.99951171875 }, { "word": "时", "start": 884.96, "end": 885.1800000000001, "probability": 0.9951171875 }, { "word": "后", "start": 885.1800000000001, "end": 885.4200000000001, "probability": 0.99609375 }, { "word": "抵", "start": 885.4200000000001, "end": 885.7600000000001, "probability": 0.9775390625 }, { "word": "达", "start": 885.7600000000001, "end": 886.1400000000001, "probability": 1.0 }, { "word": "东", "start": 886.1400000000001, "end": 886.7200000000001, "probability": 0.9931640625 }, { "word": "京", "start": 886.7200000000001, "end": 887.08, "probability": 0.994140625 }, { "word": "雨", "start": 887.08, "end": 887.58, "probability": 0.390869140625 }, { "word": "田", "start": 887.58, "end": 887.8000000000001, "probability": 0.86376953125 }, { "word": "国", "start": 887.8000000000001, "end": 888.0200000000001, "probability": 0.9990234375 }, { "word": "际", "start": 888.0200000000001, "end": 888.2, "probability": 0.98681640625 }, { "word": "机", "start": 888.2, "end": 888.3600000000001, "probability": 0.9990234375 }, { "word": "场", "start": 888.3600000000001, "end": 888.6400000000001, "probability": 0.9990234375 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "各位贵宾,飞机预计将在一个小时后抵达东京羽田国际机场", "review_status": "CERTAIN_RESTORATION", "classification": "CERTAIN_RESTORATION", "confidence": 0.95, "correction_reason": "Independent ASR confirms '羽田' (Haneda) correcting the ASR error '雨田' (Uta), consistent with flight destination." }, { "start": 889.3000000000001, "end": 893.4800000000001, "text": "当地的天气为雨天,气温为7度", "raw_asr_text": "当地的天气为雨天,气温为7度", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 53, "words": [ { "word": "当", "start": 889.3000000000001, "end": 889.7400000000001, "probability": 0.2294921875 }, { "word": "地", "start": 889.7400000000001, "end": 889.96, "probability": 1.0 }, { "word": "的", "start": 889.96, "end": 890.1000000000001, "probability": 0.9990234375 }, { "word": "天", "start": 890.1000000000001, "end": 890.2600000000001, "probability": 0.99951171875 }, { "word": "气", "start": 890.2600000000001, "end": 890.58, "probability": 1.0 }, { "word": "为", "start": 890.58, "end": 890.82, "probability": 0.99560546875 }, { "word": "雨", "start": 890.82, "end": 891.1000000000001, "probability": 0.99951171875 }, { "word": "天,", "start": 891.1000000000001, "end": 891.4200000000001, "probability": 0.99951171875 }, { "word": "气", "start": 892.0600000000001, "end": 892.1400000000001, "probability": 0.99853515625 }, { "word": "温", "start": 892.1400000000001, "end": 892.3400000000001, "probability": 0.9990234375 }, { "word": "为", "start": 892.3400000000001, "end": 892.6400000000001, "probability": 0.99951171875 }, { "word": "7", "start": 892.6400000000001, "end": 893.1600000000001, "probability": 0.6220703125 }, { "word": "度", "start": 893.1600000000001, "end": 893.4800000000001, "probability": 0.99951171875 } ], "asr_metrics": { "avg_logprob": -0.11775940981778231, "compression_ratio": 1.172093023255814, "no_speech_prob": 1.9891798277993367e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "zh", "text": "各位贵宾,飞机预计将在一个小时后抵达东京雨田国际机场当地的天气为雨天,气温为7度请您再次确认日本的入境资料是否已经填写好持有它国护照的旅客必须要填写入国登记卡以及海关申报单" }, { "language": "ja", "text": "ご視聴ありがとうございました。" } ], "id": 137, "review_annotation": "", "row_id": 137, "bounded_row_id": 137, "parent_row_ids": [ 137 ], "parent_row_id": 137, "packet_id": 53, "LID_evidence": [ { "region_id": 130, "probabilities": { "ja": 0.009319866800473796, "zh": 0.9243888523843553, "en": 0.0509151482126779 }, "status": "STABLE" }, { "region_id": 131, "probabilities": { "ja": 0.016656276944559067, "zh": 0.8712802678346634, "en": 0.0841309386305511 }, "status": "STABLE" } ], "VAD_region_ids": [ 130, 131 ], "language_confidence": 0.9243888523843553, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 880.9000000000001, "packet_end": 906.9, "raw_text": "当地的天气为雨天,气温为7度", "selected_text": "当地的天气为雨天,气温为7度", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 137 ], "packet_id": 53, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "当地的天气为雨天,气温为7度", "after": "当地的天气为雨天,气温为7度", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Text matches standard Chinese announcement phrasing and context perfectly." } ], "language": "zh", "packet_ids": [ 53 ], "index": 137, "source_text": "当地的天气为雨天,气温为7度", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Text matches standard Chinese announcement phrasing and context perfectly." }, { "start": 895.2200000000001, "end": 899.46, "text": "请您再次确认日本的入境资料是否已经填写好", "raw_asr_text": "请您再次确认日本的入境资料是否已经填写好", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 53, "words": [ { "word": "请", "start": 895.2200000000001, "end": 895.6600000000001, "probability": 0.9892578125 }, { "word": "您", "start": 895.6600000000001, "end": 895.8600000000001, "probability": 0.9990234375 }, { "word": "再", "start": 895.8600000000001, "end": 896.08, "probability": 0.9970703125 }, { "word": "次", "start": 896.08, "end": 896.2800000000001, "probability": 0.9990234375 }, { "word": "确", "start": 896.2800000000001, "end": 896.4800000000001, "probability": 1.0 }, { "word": "认", "start": 896.4800000000001, "end": 896.7400000000001, "probability": 1.0 }, { "word": "日本", "start": 896.7400000000001, "end": 897.1800000000001, "probability": 0.9296875 }, { "word": "的", "start": 897.1800000000001, "end": 897.3400000000001, "probability": 0.99755859375 }, { "word": "入", "start": 897.3400000000001, "end": 897.5000000000001, "probability": 0.9990234375 }, { "word": "境", "start": 897.5000000000001, "end": 897.7, "probability": 0.98095703125 }, { "word": "资", "start": 897.7, "end": 897.9200000000001, "probability": 0.9990234375 }, { "word": "料", "start": 897.9200000000001, "end": 898.08, "probability": 0.99951171875 }, { "word": "是", "start": 898.08, "end": 898.3400000000001, "probability": 0.99951171875 }, { "word": "否", "start": 898.3400000000001, "end": 898.46, "probability": 1.0 }, { "word": "已经", "start": 898.46, "end": 898.7600000000001, "probability": 0.99560546875 }, { "word": "填", "start": 898.7600000000001, "end": 898.94, "probability": 0.99609375 }, { "word": "写", "start": 898.94, "end": 899.1600000000001, "probability": 0.99951171875 }, { "word": "好", "start": 899.1600000000001, "end": 899.46, "probability": 0.99609375 } ], "asr_metrics": { "avg_logprob": -0.11775940981778231, "compression_ratio": 1.172093023255814, "no_speech_prob": 1.9891798277993367e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "zh", "text": "各位贵宾,飞机预计将在一个小时后抵达东京雨田国际机场当地的天气为雨天,气温为7度请您再次确认日本的入境资料是否已经填写好持有它国护照的旅客必须要填写入国登记卡以及海关申报单" }, { "language": "ja", "text": "ご視聴ありがとうございました。" } ], "id": 138, "review_annotation": "", "row_id": 138, "bounded_row_id": 138, "parent_row_ids": [ 138 ], "parent_row_id": 138, "packet_id": 53, "LID_evidence": [ { "region_id": 132, "probabilities": { "ja": 0.0032658301590179854, "zh": 0.9737254462458871, "en": 0.016809362002690745 }, "status": "STABLE" } ], "VAD_region_ids": [ 132 ], "language_confidence": 0.9737254462458871, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 880.9000000000001, "packet_end": 906.9, "raw_text": "请您再次确认日本的入境资料是否已经填写好", "selected_text": "请您再次确认日本的入境资料是否已经填写好", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 138 ], "packet_id": 53, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "请您再次确认日本的入境资料是否已经填写好", "after": "请您再次确认日本的入境资料是否已经填写好", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Text matches standard Chinese announcement phrasing and context perfectly." } ], "language": "zh", "packet_ids": [ 53 ], "index": 138, "source_text": "请您再次确认日本的入境资料是否已经填写好", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Text matches standard Chinese announcement phrasing and context perfectly." }, { "start": 899.46, "end": 905.7600000000001, "text": "持有它国护照的旅客必须要填写入国登记卡以及海关申报单", "raw_asr_text": "持有它国护照的旅客必须要填写入国登记卡以及海关申报单", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 53, "words": [ { "word": "持", "start": 899.46, "end": 900.2800000000001, "probability": 0.91162109375 }, { "word": "有", "start": 900.2800000000001, "end": 900.4200000000001, "probability": 0.99951171875 }, { "word": "它", "start": 900.4200000000001, "end": 900.58, "probability": 0.5625 }, { "word": "国", "start": 900.58, "end": 900.7800000000001, "probability": 0.99609375 }, { "word": "护", "start": 900.7800000000001, "end": 900.94, "probability": 0.9775390625 }, { "word": "照", "start": 900.94, "end": 901.1200000000001, "probability": 0.9912109375 }, { "word": "的", "start": 901.1200000000001, "end": 901.2600000000001, "probability": 0.99951171875 }, { "word": "旅", "start": 901.2600000000001, "end": 901.44, "probability": 0.99951171875 }, { "word": "客", "start": 901.44, "end": 901.6400000000001, "probability": 0.9931640625 }, { "word": "必", "start": 901.6400000000001, "end": 901.9000000000001, "probability": 0.99951171875 }, { "word": "须", "start": 901.9000000000001, "end": 902.08, "probability": 0.99658203125 }, { "word": "要", "start": 902.08, "end": 902.2200000000001, "probability": 0.99267578125 }, { "word": "填", "start": 902.2200000000001, "end": 902.46, "probability": 0.99951171875 }, { "word": "写", "start": 902.46, "end": 902.8600000000001, "probability": 1.0 }, { "word": "入", "start": 902.8600000000001, "end": 903.46, "probability": 0.5791015625 }, { "word": "国", "start": 903.46, "end": 903.58, "probability": 0.982421875 }, { "word": "登", "start": 903.58, "end": 903.7800000000001, "probability": 0.9833984375 }, { "word": "记", "start": 903.7800000000001, "end": 903.9800000000001, "probability": 0.94873046875 }, { "word": "卡", "start": 903.9800000000001, "end": 904.3400000000001, "probability": 0.99853515625 }, { "word": "以及", "start": 904.3400000000001, "end": 904.6600000000001, "probability": 0.97412109375 }, { "word": "海", "start": 904.6600000000001, "end": 904.9000000000001, "probability": 0.99853515625 }, { "word": "关", "start": 904.9000000000001, "end": 905.1400000000001, "probability": 0.99658203125 }, { "word": "申", "start": 905.1400000000001, "end": 905.32, "probability": 0.9931640625 }, { "word": "报", "start": 905.32, "end": 905.5000000000001, "probability": 0.984375 }, { "word": "单", "start": 905.5000000000001, "end": 905.7600000000001, "probability": 0.998046875 } ], "asr_metrics": { "avg_logprob": -0.11775940981778231, "compression_ratio": 1.172093023255814, "no_speech_prob": 1.9891798277993367e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "zh", "text": "各位贵宾,飞机预计将在一个小时后抵达东京雨田国际机场当地的天气为雨天,气温为7度请您再次确认日本的入境资料是否已经填写好持有它国护照的旅客必须要填写入国登记卡以及海关申报单" }, { "language": "ja", "text": "ご視聴ありがとうございました。" } ], "id": 139, "review_annotation": "", "row_id": 139, "bounded_row_id": 139, "parent_row_ids": [ 139 ], "parent_row_id": 139, "packet_id": 53, "LID_evidence": [ { "region_id": 132, "probabilities": { "ja": 0.0032658301590179854, "zh": 0.9737254462458871, "en": 0.016809362002690745 }, "status": "STABLE" } ], "VAD_region_ids": [ 132 ], "language_confidence": 0.9737254462458871, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 880.9000000000001, "packet_end": 906.9, "raw_text": "持有它国护照的旅客必须要填写入国登记卡以及海关申报单", "selected_text": "持有它国护照的旅客必须要填写入国登记卡以及海关申报单", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 139 ], "packet_id": 53, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "持有它国护照的旅客必须要填写入国登记卡以及海关申报单", "after": "持有它国护照的旅客必须要填写入国登记卡以及海关申报单", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Text matches standard Chinese announcement phrasing and context perfectly." } ], "language": "zh", "packet_ids": [ 53 ], "index": 139, "source_text": "持有它国护照的旅客必须要填写入国登记卡以及海关申报单", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Text matches standard Chinese announcement phrasing and context perfectly." }, { "start": 906.52, "end": 909.58, "text": "有需要的貴賓請與空服人員索取", "raw_asr_text": "有需要的貴賓請與空服人員索取", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 54, "words": [ { "word": "有", "start": 906.52, "end": 906.9200000000001, "probability": 0.7353515625 }, { "word": "需要", "start": 906.9200000000001, "end": 907.1800000000001, "probability": 0.994140625 }, { "word": "的", "start": 907.1800000000001, "end": 907.38, "probability": 0.99755859375 }, { "word": "貴", "start": 907.38, "end": 907.5400000000001, "probability": 0.85888671875 }, { "word": "賓", "start": 907.5400000000001, "end": 907.76, "probability": 0.9951171875 }, { "word": "請", "start": 907.76, "end": 908.02, "probability": 0.87255859375 }, { "word": "與", "start": 908.02, "end": 908.24, "probability": 0.73095703125 }, { "word": "空", "start": 908.24, "end": 908.5, "probability": 0.93310546875 }, { "word": "服", "start": 908.5, "end": 908.66, "probability": 0.63037109375 }, { "word": "人", "start": 908.66, "end": 908.84, "probability": 0.9912109375 }, { "word": "員", "start": 908.84, "end": 909.0600000000001, "probability": 0.99658203125 }, { "word": "索", "start": 909.0600000000001, "end": 909.3000000000001, "probability": 0.982421875 }, { "word": "取", "start": 909.3000000000001, "end": 909.58, "probability": 0.99951171875 } ], "asr_metrics": { "avg_logprob": -0.15825584530830383, "compression_ratio": 1.2473118279569892, "no_speech_prob": 2.4580018576081386e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "有需要的貴賓請與空服人員索取收到機場指示,稍後飛機將準備下降想要使用洗手間的貴賓,請您盡早使用洗手間當安全帶指示燈亮起時,請您儘快回到座位並確實系好安全帶,謝謝" } ], "id": 140, "review_annotation": "", "row_id": 140, "bounded_row_id": 140, "parent_row_ids": [ 140 ], "parent_row_id": 140, "packet_id": 54, "LID_evidence": [ { "region_id": 133, "probabilities": { "ja": 0.003903751925099641, "zh": 0.9660810649394989, "en": 0.021486646286211906 }, "status": "STABLE" } ], "VAD_region_ids": [ 133 ], "language_confidence": 0.9660810649394989, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 905.6, "packet_end": 926.1999999999999, "raw_text": "有需要的貴賓請與空服人員索取", "selected_text": "有需要的貴賓請與空服人員索取", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 140 ], "packet_id": 54, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "有需要的貴賓請與空服人員索取", "after": "有需要的貴賓請與空服人員索取", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard phrase for requesting forms; independent ASR confirms meaning despite minor transcription noise." } ], "language": "zh", "packet_ids": [ 54 ], "index": 140, "independent_ASR": { "text": "三，有需要的贵宾，请与空服人员索取。收到机长指示，稍后飞机将准备下降。想要使用洗手间的贵宾，请您尽早使用洗手间。当安全带指示灯亮起时，请您尽快回到座位，并确实系好安全带。谢谢。", "chunks": [ { "text": "三，有需要的贵宾，请与空服人员索取。收到机长指示，稍后飞机将准备下降。想要使用洗手间的贵宾，请您尽早使用洗手间。当安全带指示灯亮起时，请您尽快回到座位，并确实系好安全带。谢谢。", "start": 0.0, "end": 20.6, "chunk_index": 0, "language": "Chinese", "finish_reason": "eos", "truncated": false, "generated_tokens": 59, "max_new_tokens": 280 } ], "detected_language": "Chinese", "segments": [ { "start": 905.6, "end": 926.2, "text": "三，有需要的贵宾，请与空服人员索取。收到机长指示，稍后飞机将准备下降。想要使用洗手间的贵宾，请您尽早使用洗手间。当安全带指示灯亮起时，请您尽快回到座位，并确实系好安全带。谢谢。" } ], "start": 905.6, "end": 926.1999999999999, "engine": "mlx-qwen3-asr", "language": "zh" }, "source_text": "有需要的貴賓請與空服人員索取", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.9, "correction_reason": "Standard phrase for requesting forms; independent ASR confirms meaning despite minor transcription noise." }, { "start": 910.5600000000001, "end": 913.52, "text": "收到機長指示,稍後飛機將準備下降", "raw_asr_text": "收到機場指示,稍後飛機將準備下降", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 54, "words": [], "asr_metrics": { "avg_logprob": -0.15825584530830383, "compression_ratio": 1.2473118279569892, "no_speech_prob": 2.4580018576081386e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "有需要的貴賓請與空服人員索取收到機場指示,稍後飛機將準備下降想要使用洗手間的貴賓,請您盡早使用洗手間當安全帶指示燈亮起時,請您儘快回到座位並確實系好安全帶,謝謝" } ], "id": 141, "review_annotation": "", "row_id": 141, "bounded_row_id": 141, "parent_row_ids": [ 141 ], "parent_row_id": 141, "packet_id": 54, "LID_evidence": [ { "region_id": 134, "probabilities": { "ja": 0.012304250142187811, "zh": 0.8644777335226536, "en": 0.028804436139762402 }, "status": "STABLE" } ], "VAD_region_ids": [ 134 ], "language_confidence": 0.8644777335226536, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 905.6, "packet_end": 926.1999999999999, "raw_text": "收到機場指示,稍後飛機將準備下降", "selected_text": "收到機長指示,稍後飛機將準備下降", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 141 ], "packet_id": 54, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "收到機場指示,稍後飛機將準備下降", "after": "收到機長指示,稍後飛機將準備下降", "classification": "CERTAIN_RESTORATION", "reason": "Independent ASR confirms '機長' (Captain) correcting '機場' (Airport), fitting the context of descent preparation." } ], "language": "zh", "packet_ids": [ 54 ], "index": 141, "independent_ASR": { "text": "三，有需要的贵宾，请与空服人员索取。收到机长指示，稍后飞机将准备下降。想要使用洗手间的贵宾，请您尽早使用洗手间。当安全带指示灯亮起时，请您尽快回到座位，并确实系好安全带。谢谢。", "chunks": [ { "text": "三，有需要的贵宾，请与空服人员索取。收到机长指示，稍后飞机将准备下降。想要使用洗手间的贵宾，请您尽早使用洗手间。当安全带指示灯亮起时，请您尽快回到座位，并确实系好安全带。谢谢。", "start": 0.0, "end": 20.6, "chunk_index": 0, "language": "Chinese", "finish_reason": "eos", "truncated": false, "generated_tokens": 59, "max_new_tokens": 280 } ], "detected_language": "Chinese", "segments": [ { "start": 905.6, "end": 926.2, "text": "三，有需要的贵宾，请与空服人员索取。收到机长指示，稍后飞机将准备下降。想要使用洗手间的贵宾，请您尽早使用洗手间。当安全带指示灯亮起时，请您尽快回到座位，并确实系好安全带。谢谢。" } ], "start": 905.6, "end": 926.1999999999999, "engine": "mlx-qwen3-asr", "language": "zh" }, "raw_word_timestamps": [ { "word": "收", "start": 910.5600000000001, "end": 910.96, "probability": 0.888671875 }, { "word": "到", "start": 910.96, "end": 911.1, "probability": 0.998046875 }, { "word": "機", "start": 911.1, "end": 911.24, "probability": 0.98291015625 }, { "word": "場", "start": 911.24, "end": 911.4, "probability": 0.66552734375 }, { "word": "指", "start": 911.4, "end": 911.58, "probability": 0.99365234375 }, { "word": "示,", "start": 911.58, "end": 911.8000000000001, "probability": 0.99853515625 }, { "word": "稍", "start": 911.94, "end": 912.0400000000001, "probability": 0.990234375 }, { "word": "後", "start": 912.0400000000001, "end": 912.24, "probability": 0.9931640625 }, { "word": "飛", "start": 912.24, "end": 912.38, "probability": 0.955078125 }, { "word": "機", "start": 912.38, "end": 912.5400000000001, "probability": 0.994140625 }, { "word": "將", "start": 912.5400000000001, "end": 912.6800000000001, "probability": 0.9765625 }, { "word": "準備", "start": 912.6800000000001, "end": 912.9, "probability": 0.99462890625 }, { "word": "下", "start": 912.9, "end": 913.24, "probability": 0.99658203125 }, { "word": "降", "start": 913.24, "end": 913.52, "probability": 0.99853515625 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "收到機長指示,稍後飛機將準備下降", "review_status": "CERTAIN_RESTORATION", "classification": "CERTAIN_RESTORATION", "confidence": 0.95, "correction_reason": "Independent ASR confirms '機長' (Captain) correcting '機場' (Airport), fitting the context of descent preparation." }, { "start": 914.46, "end": 918.12, "text": "想要使用洗手間的貴賓,請您盡早使用洗手間", "raw_asr_text": "想要使用洗手間的貴賓,請您盡早使用洗手間", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 54, "words": [ { "word": "想", "start": 914.46, "end": 914.86, "probability": 0.95654296875 }, { "word": "要", "start": 914.86, "end": 915.02, "probability": 0.98681640625 }, { "word": "使", "start": 915.02, "end": 915.14, "probability": 0.80810546875 }, { "word": "用", "start": 915.14, "end": 915.34, "probability": 0.99951171875 }, { "word": "洗", "start": 915.34, "end": 915.5, "probability": 0.998046875 }, { "word": "手", "start": 915.5, "end": 915.6800000000001, "probability": 1.0 }, { "word": "間", "start": 915.6800000000001, "end": 915.82, "probability": 0.99072265625 }, { "word": "的", "start": 915.82, "end": 915.98, "probability": 0.9970703125 }, { "word": "貴", "start": 915.98, "end": 916.16, "probability": 0.99951171875 }, { "word": "賓,", "start": 916.16, "end": 916.34, "probability": 1.0 }, { "word": "請", "start": 916.4200000000001, "end": 916.5600000000001, "probability": 0.9951171875 }, { "word": "您", "start": 916.5600000000001, "end": 916.74, "probability": 0.9951171875 }, { "word": "盡", "start": 916.74, "end": 916.98, "probability": 0.76123046875 }, { "word": "早", "start": 916.98, "end": 917.16, "probability": 0.9970703125 }, { "word": "使", "start": 917.16, "end": 917.4, "probability": 0.986328125 }, { "word": "用", "start": 917.4, "end": 917.5400000000001, "probability": 0.9990234375 }, { "word": "洗", "start": 917.5400000000001, "end": 917.7, "probability": 0.99853515625 }, { "word": "手", "start": 917.7, "end": 917.9, "probability": 1.0 }, { "word": "間", "start": 917.9, "end": 918.12, "probability": 0.99658203125 } ], "asr_metrics": { "avg_logprob": -0.15825584530830383, "compression_ratio": 1.2473118279569892, "no_speech_prob": 2.4580018576081386e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "有需要的貴賓請與空服人員索取收到機場指示,稍後飛機將準備下降想要使用洗手間的貴賓,請您盡早使用洗手間當安全帶指示燈亮起時,請您儘快回到座位並確實系好安全帶,謝謝" } ], "id": 142, "review_annotation": "", "row_id": 142, "bounded_row_id": 142, "parent_row_ids": [ 142 ], "parent_row_id": 142, "packet_id": 54, "LID_evidence": [ { "region_id": 135, "probabilities": { "ja": 0.008817285988091802, "zh": 0.9012728463858366, "en": 0.01567283961533879 }, "status": "STABLE" } ], "VAD_region_ids": [ 135 ], "language_confidence": 0.9012728463858366, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 905.6, "packet_end": 926.1999999999999, "raw_text": "想要使用洗手間的貴賓,請您盡早使用洗手間", "selected_text": "想要使用洗手間的貴賓,請您盡早使用洗手間", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 142 ], "packet_id": 54, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "想要使用洗手間的貴賓,請您盡早使用洗手間", "after": "想要使用洗手間的貴賓,請您盡早使用洗手間", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard announcement phrasing; independent ASR confirms the request to use lavatories early." } ], "language": "zh", "packet_ids": [ 54 ], "index": 142, "independent_ASR": { "text": "三，有需要的贵宾，请与空服人员索取。收到机长指示，稍后飞机将准备下降。想要使用洗手间的贵宾，请您尽早使用洗手间。当安全带指示灯亮起时，请您尽快回到座位，并确实系好安全带。谢谢。", "chunks": [ { "text": "三，有需要的贵宾，请与空服人员索取。收到机长指示，稍后飞机将准备下降。想要使用洗手间的贵宾，请您尽早使用洗手间。当安全带指示灯亮起时，请您尽快回到座位，并确实系好安全带。谢谢。", "start": 0.0, "end": 20.6, "chunk_index": 0, "language": "Chinese", "finish_reason": "eos", "truncated": false, "generated_tokens": 59, "max_new_tokens": 280 } ], "detected_language": "Chinese", "segments": [ { "start": 905.6, "end": 926.2, "text": "三，有需要的贵宾，请与空服人员索取。收到机长指示，稍后飞机将准备下降。想要使用洗手间的贵宾，请您尽早使用洗手间。当安全带指示灯亮起时，请您尽快回到座位，并确实系好安全带。谢谢。" } ], "start": 905.6, "end": 926.1999999999999, "engine": "mlx-qwen3-asr", "language": "zh" }, "source_text": "想要使用洗手間的貴賓,請您盡早使用洗手間", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.9, "correction_reason": "Standard announcement phrasing; independent ASR confirms the request to use lavatories early." }, { "start": 918.6, "end": 922.5600000000001, "text": "當安全帶指示燈亮起時,請您儘快回到座位", "raw_asr_text": "當安全帶指示燈亮起時,請您儘快回到座位", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 54, "words": [ { "word": "當", "start": 918.6, "end": 919.0, "probability": 0.958984375 }, { "word": "安全", "start": 919.0, "end": 919.36, "probability": 0.99609375 }, { "word": "帶", "start": 919.36, "end": 919.5400000000001, "probability": 0.93994140625 }, { "word": "指", "start": 919.5400000000001, "end": 919.76, "probability": 0.9189453125 }, { "word": "示", "start": 919.76, "end": 919.9200000000001, "probability": 0.98828125 }, { "word": "燈", "start": 919.9200000000001, "end": 920.1, "probability": 0.947265625 }, { "word": "亮", "start": 920.1, "end": 920.28, "probability": 0.9853515625 }, { "word": "起", "start": 920.28, "end": 920.5600000000001, "probability": 0.99267578125 }, { "word": "時,", "start": 920.5600000000001, "end": 920.76, "probability": 0.97607421875 }, { "word": "請", "start": 920.9200000000001, "end": 921.0600000000001, "probability": 0.9931640625 }, { "word": "您", "start": 921.0600000000001, "end": 921.28, "probability": 0.99462890625 }, { "word": "儘", "start": 921.28, "end": 921.52, "probability": 0.791015625 }, { "word": "快", "start": 921.52, "end": 921.74, "probability": 0.99462890625 }, { "word": "回", "start": 921.74, "end": 921.9200000000001, "probability": 0.98779296875 }, { "word": "到", "start": 921.9200000000001, "end": 922.12, "probability": 0.9990234375 }, { "word": "座", "start": 922.12, "end": 922.34, "probability": 0.994140625 }, { "word": "位", "start": 922.34, "end": 922.5600000000001, "probability": 1.0 } ], "asr_metrics": { "avg_logprob": -0.15825584530830383, "compression_ratio": 1.2473118279569892, "no_speech_prob": 2.4580018576081386e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "有需要的貴賓請與空服人員索取收到機場指示,稍後飛機將準備下降想要使用洗手間的貴賓,請您盡早使用洗手間當安全帶指示燈亮起時,請您儘快回到座位並確實系好安全帶,謝謝" } ], "id": 143, "review_annotation": "", "row_id": 143, "bounded_row_id": 143, "parent_row_ids": [ 143 ], "parent_row_id": 143, "packet_id": 54, "LID_evidence": [ { "region_id": 136, "probabilities": { "ja": 0.0018085615842470101, "zh": 0.9857037961483002, "en": 0.010076845530420542 }, "status": "STABLE" } ], "VAD_region_ids": [ 136 ], "language_confidence": 0.9857037961483002, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 905.6, "packet_end": 926.1999999999999, "raw_text": "當安全帶指示燈亮起時,請您儘快回到座位", "selected_text": "當安全帶指示燈亮起時,請您儘快回到座位", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 143 ], "packet_id": 54, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "當安全帶指示燈亮起時,請您儘快回到座位", "after": "當安全帶指示燈亮起時,請您儘快回到座位", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard safety instruction; independent ASR confirms the instruction to return to seats when the sign is on." } ], "language": "zh", "packet_ids": [ 54 ], "index": 143, "independent_ASR": { "text": "三，有需要的贵宾，请与空服人员索取。收到机长指示，稍后飞机将准备下降。想要使用洗手间的贵宾，请您尽早使用洗手间。当安全带指示灯亮起时，请您尽快回到座位，并确实系好安全带。谢谢。", "chunks": [ { "text": "三，有需要的贵宾，请与空服人员索取。收到机长指示，稍后飞机将准备下降。想要使用洗手间的贵宾，请您尽早使用洗手间。当安全带指示灯亮起时，请您尽快回到座位，并确实系好安全带。谢谢。", "start": 0.0, "end": 20.6, "chunk_index": 0, "language": "Chinese", "finish_reason": "eos", "truncated": false, "generated_tokens": 59, "max_new_tokens": 280 } ], "detected_language": "Chinese", "segments": [ { "start": 905.6, "end": 926.2, "text": "三，有需要的贵宾，请与空服人员索取。收到机长指示，稍后飞机将准备下降。想要使用洗手间的贵宾，请您尽早使用洗手间。当安全带指示灯亮起时，请您尽快回到座位，并确实系好安全带。谢谢。" } ], "start": 905.6, "end": 926.1999999999999, "engine": "mlx-qwen3-asr", "language": "zh" }, "source_text": "當安全帶指示燈亮起時,請您儘快回到座位", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.9, "correction_reason": "Standard safety instruction; independent ASR confirms the instruction to return to seats when the sign is on." }, { "start": 922.5600000000001, "end": 924.9, "text": "並確實系好安全帶,謝謝", "raw_asr_text": "並確實系好安全帶,謝謝", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 54, "words": [ { "word": "並", "start": 922.5600000000001, "end": 922.86, "probability": 0.69677734375 }, { "word": "確", "start": 922.86, "end": 923.08, "probability": 0.97998046875 }, { "word": "實", "start": 923.08, "end": 923.36, "probability": 0.9892578125 }, { "word": "系", "start": 923.36, "end": 923.48, "probability": 0.2442626953125 }, { "word": "好", "start": 923.48, "end": 923.74, "probability": 0.99755859375 }, { "word": "安全", "start": 923.74, "end": 924.0400000000001, "probability": 0.99658203125 }, { "word": "帶,", "start": 924.0400000000001, "end": 924.38, "probability": 0.99853515625 }, { "word": "謝謝", "start": 924.62, "end": 924.9, "probability": 0.990234375 } ], "asr_metrics": { "avg_logprob": -0.15825584530830383, "compression_ratio": 1.2473118279569892, "no_speech_prob": 2.4580018576081386e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "有需要的貴賓請與空服人員索取收到機場指示,稍後飛機將準備下降想要使用洗手間的貴賓,請您盡早使用洗手間當安全帶指示燈亮起時,請您儘快回到座位並確實系好安全帶,謝謝" } ], "id": 144, "review_annotation": "", "row_id": 144, "bounded_row_id": 144, "parent_row_ids": [ 144 ], "parent_row_id": 144, "packet_id": 54, "LID_evidence": [ { "region_id": 136, "probabilities": { "ja": 0.0018085615842470101, "zh": 0.9857037961483002, "en": 0.010076845530420542 }, "status": "STABLE" } ], "VAD_region_ids": [ 136 ], "language_confidence": 0.9857037961483002, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 905.6, "packet_end": 926.1999999999999, "raw_text": "並確實系好安全帶,謝謝", "selected_text": "並確實系好安全帶,謝謝", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 144 ], "packet_id": 54, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "並確實系好安全帶,謝謝", "after": "並確實系好安全帶,謝謝", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard closing for safety instruction; independent ASR confirms the request to fasten seatbelts." } ], "language": "zh", "packet_ids": [ 54 ], "index": 144, "independent_ASR": { "text": "三，有需要的贵宾，请与空服人员索取。收到机长指示，稍后飞机将准备下降。想要使用洗手间的贵宾，请您尽早使用洗手间。当安全带指示灯亮起时，请您尽快回到座位，并确实系好安全带。谢谢。", "chunks": [ { "text": "三，有需要的贵宾，请与空服人员索取。收到机长指示，稍后飞机将准备下降。想要使用洗手间的贵宾，请您尽早使用洗手间。当安全带指示灯亮起时，请您尽快回到座位，并确实系好安全带。谢谢。", "start": 0.0, "end": 20.6, "chunk_index": 0, "language": "Chinese", "finish_reason": "eos", "truncated": false, "generated_tokens": 59, "max_new_tokens": 280 } ], "detected_language": "Chinese", "segments": [ { "start": 905.6, "end": 926.2, "text": "三，有需要的贵宾，请与空服人员索取。收到机长指示，稍后飞机将准备下降。想要使用洗手间的贵宾，请您尽早使用洗手间。当安全带指示灯亮起时，请您尽快回到座位，并确实系好安全带。谢谢。" } ], "start": 905.6, "end": 926.1999999999999, "engine": "mlx-qwen3-asr", "language": "zh" }, "source_text": "並確實系好安全帶,謝謝", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.9, "correction_reason": "Standard closing for safety instruction; independent ASR confirms the request to fasten seatbelts." }, { "start": 928.0, "end": 931.9200000000001, "text": "皆様、着陸準備のため、免税品販売を終了いたします。", "raw_asr_text": "皆様、着陸準備のため、免税品販売を終了いたします。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 55, "words": [], "asr_metrics": { "avg_logprob": -0.14214953860721072, "compression_ratio": 1.1150442477876106, "no_speech_prob": 6.07770864885504e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "皆様、着陸準備のため、免税品販売を終了いたします。お買い上げありがとうございました。" } ], "id": 145, "review_annotation": "", "row_id": 145, "bounded_row_id": 145, "parent_row_ids": [ 145 ], "parent_row_id": 145, "packet_id": 55, "LID_evidence": [ { "region_id": 137, "probabilities": { "ja": 0.9901072382926941, "zh": 0.0007392901388811879, "en": 0.004187123622978106 }, "status": "STABLE" } ], "VAD_region_ids": [ 137 ], "language_confidence": 0.9901072382926941, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 927.2, "packet_end": 935.3, "raw_text": "皆様、着陸準備のため、免税品販売を終了いたします。", "selected_text": "皆様、着陸準備のため、免税品販売を終了いたします。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 145 ], "packet_id": 55, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "皆様、着陸準備のため、免税品販売を終了いたします。", "after": "皆様、着陸準備のため、免税品販売を終了いたします。", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard Japanese announcement. No ASR errors detected." } ], "language": "ja", "packet_ids": [ 55 ], "index": 145, "source_text": "皆様、着陸準備のため、免税品販売を終了いたします。", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard Japanese announcement. No ASR errors detected.", "raw_word_timestamps": [ { "word": "皆", "start": 927.94, "end": 928.32, "probability": 0.97802734375 }, { "word": "様、", "start": 928.32, "end": 928.5400000000001, "probability": 0.662109375 }, { "word": "着", "start": 928.6800000000001, "end": 928.82, "probability": 1.0 }, { "word": "陸", "start": 928.82, "end": 929.0600000000001, "probability": 1.0 }, { "word": "準備", "start": 929.0600000000001, "end": 929.38, "probability": 1.0 }, { "word": "の", "start": 929.38, "end": 929.6, "probability": 1.0 }, { "word": "ため、", "start": 929.6, "end": 929.88, "probability": 1.0 }, { "word": "免", "start": 930.0400000000001, "end": 930.1600000000001, "probability": 0.9765625 }, { "word": "税", "start": 930.1600000000001, "end": 930.34, "probability": 0.990234375 }, { "word": "品", "start": 930.34, "end": 930.5, "probability": 0.98876953125 }, { "word": "販", "start": 930.5, "end": 930.76, "probability": 1.0 }, { "word": "売", "start": 930.76, "end": 930.96, "probability": 1.0 }, { "word": "を", "start": 930.96, "end": 931.08, "probability": 1.0 }, { "word": "終", "start": 931.08, "end": 931.26, "probability": 0.99853515625 }, { "word": "了", "start": 931.26, "end": 931.46, "probability": 1.0 }, { "word": "いた", "start": 931.46, "end": 931.6, "probability": 0.99609375 }, { "word": "します。", "start": 931.6, "end": 931.9200000000001, "probability": 1.0 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 932.1800000000001, "end": 933.3000000000001, "text": "お買い上げありがとうございました。", "raw_asr_text": "お買い上げありがとうございました。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 55, "words": [ { "word": "お", "start": 932.1800000000001, "end": 932.44, "probability": 0.9970703125 }, { "word": "買", "start": 932.44, "end": 932.58, "probability": 0.9990234375 }, { "word": "い", "start": 932.58, "end": 932.6600000000001, "probability": 0.99951171875 }, { "word": "上", "start": 932.6600000000001, "end": 932.74, "probability": 0.9970703125 }, { "word": "げ", "start": 932.74, "end": 932.86, "probability": 1.0 }, { "word": "ありがとうございました。", "start": 932.86, "end": 933.3000000000001, "probability": 0.974609375 } ], "asr_metrics": { "avg_logprob": -0.14214953860721072, "compression_ratio": 1.1150442477876106, "no_speech_prob": 6.07770864885504e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "皆様、着陸準備のため、免税品販売を終了いたします。お買い上げありがとうございました。" } ], "id": 146, "review_annotation": "", "row_id": 146, "bounded_row_id": 146, "parent_row_ids": [ 146 ], "parent_row_id": 146, "packet_id": 55, "LID_evidence": [ { "region_id": 137, "probabilities": { "ja": 0.9901072382926941, "zh": 0.0007392901388811879, "en": 0.004187123622978106 }, "status": "STABLE" } ], "VAD_region_ids": [ 137 ], "language_confidence": 0.9901072382926941, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 927.2, "packet_end": 935.3, "raw_text": "お買い上げありがとうございました。", "selected_text": "お買い上げありがとうございました。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 146 ], "packet_id": 55, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "お買い上げありがとうございました。", "after": "お買い上げありがとうございました。", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard Japanese closing phrase. No ASR errors detected." } ], "language": "ja", "packet_ids": [ 55 ], "index": 146, "source_text": "お買い上げありがとうございました。", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard Japanese closing phrase. No ASR errors detected." }, { "start": 937.4, "end": 941.86, "text": "Ladies and gentlemen, we have just closed the intermission. Thank you for your purchase.", "raw_asr_text": "Ladies and gentlemen, we have just closed inter-says. Thank you for your purchase.", "source_language": "en", "source_lang": "en", "asr_packet_id": 56, "words": [], "asr_metrics": { "avg_logprob": -0.38658319820057263, "compression_ratio": 1.0379746835443038, "no_speech_prob": 7.11860223545635e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "en", "text": "Ladies and gentlemen, we have just closed inter-says. Thank you for your purchase." } ], "id": 147, "review_annotation": "待確認", "row_id": 147, "bounded_row_id": 147, "parent_row_ids": [ 147 ], "parent_row_id": 147, "packet_id": 56, "LID_evidence": [ { "region_id": 138, "probabilities": { "ja": 0.028263205329754522, "zh": 0.0023682722821831703, "en": 0.916819589478629 }, "status": "STABLE" } ], "VAD_region_ids": [ 138 ], "language_confidence": 0.916819589478629, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 936.6, "packet_end": 943.0999999999999, "raw_text": "Ladies and gentlemen, we have just closed inter-says. Thank you for your purchase.", "selected_text": "Ladies and gentlemen, we have just closed the intermission. Thank you for your purchase.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 147 ], "packet_id": 56, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "Ladies and gentlemen, we have just closed inter-says. Thank you for your purchase.", "after": "Ladies and gentlemen, we have just closed the intermission. Thank you for your purchase.", "classification": "BEST_SUPPORTED_WITH_REVIEW", "reason": "Independent ASR says 'intercess' (likely 'intermission' or 'in-flight sales'); context implies sales end. Marked for review." } ], "language": "en", "packet_ids": [ 56 ], "index": 147, "independent_ASR": { "text": "Ladies and gentlemen, we have just closed the intercess. Thank you for your purchase.", "chunks": [ { "text": "Ladies and gentlemen, we have just closed the intercess. Thank you for your purchase.", "start": 0.0, "end": 6.5, "chunk_index": 0, "language": "English", "finish_reason": "eos", "truncated": false, "generated_tokens": 19, "max_new_tokens": 128 } ], "detected_language": "English", "segments": [ { "start": 936.6, "end": 943.1, "text": "Ladies and gentlemen, we have just closed the intercess. Thank you for your purchase." } ], "start": 936.6, "end": 943.0999999999999, "engine": "mlx-qwen3-asr", "language": "en" }, "raw_word_timestamps": [ { "word": " Ladies", "start": 937.22, "end": 937.78, "probability": 0.61181640625 }, { "word": " and", "start": 937.78, "end": 938.02, "probability": 0.9833984375 }, { "word": " gentlemen,", "start": 938.02, "end": 938.34, "probability": 0.7861328125 }, { "word": " we", "start": 938.5, "end": 938.58, "probability": 0.98095703125 }, { "word": " have", "start": 938.58, "end": 938.78, "probability": 0.9873046875 }, { "word": " just", "start": 938.78, "end": 939.0600000000001, "probability": 0.990234375 }, { "word": " closed", "start": 939.0600000000001, "end": 939.52, "probability": 0.97509765625 }, { "word": " inter", "start": 939.52, "end": 939.9, "probability": 0.493408203125 }, { "word": "-says.", "start": 939.9, "end": 940.46, "probability": 0.564453125 }, { "word": " Thank", "start": 940.64, "end": 940.94, "probability": 0.9912109375 }, { "word": " you", "start": 940.94, "end": 941.1, "probability": 0.98095703125 }, { "word": " for", "start": 941.1, "end": 941.3000000000001, "probability": 0.99609375 }, { "word": " your", "start": 941.3000000000001, "end": 941.46, "probability": 0.9921875 }, { "word": " purchase.", "start": 941.46, "end": 941.86, "probability": 0.97998046875 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "Ladies and gentlemen, we have just closed the intermission. Thank you for your purchase.", "review_status": "BEST_SUPPORTED_WITH_REVIEW", "classification": "BEST_SUPPORTED_WITH_REVIEW", "confidence": 0.85, "correction_reason": "Independent ASR says 'intercess' (likely 'intermission' or 'in-flight sales'); context implies sales end. Marked for review." }, { "start": 945.4, "end": 948.08, "text": "各位贵宾我们现在结束免税商品的销售", "raw_asr_text": "各位贵民我们现在结束棉市上面的销售", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 57, "words": [], "asr_metrics": { "avg_logprob": -0.36127067434376686, "compression_ratio": 0.8625, "no_speech_prob": 1.360169268882494e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "各位贵民我们现在结束棉市上面的销售感谢你的购买" } ], "id": 148, "review_annotation": "", "row_id": 148, "bounded_row_id": 148, "parent_row_ids": [ 148 ], "parent_row_id": 148, "packet_id": 57, "LID_evidence": [ { "region_id": 139, "probabilities": { "ja": 0.012560169550124557, "zh": 0.8904412806034088, "en": 0.028701536101289095 }, "status": "STABLE" } ], "VAD_region_ids": [ 139 ], "language_confidence": 0.8904412806034088, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 944.6, "packet_end": 950.3, "raw_text": "各位贵民我们现在结束棉市上面的销售", "selected_text": "各位贵宾我们现在结束免税商品的销售", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 148 ], "packet_id": 57, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "各位贵民我们现在结束棉市上面的销售", "after": "各位贵宾我们现在结束免税商品的销售", "classification": "CERTAIN_RESTORATION", "reason": "Independent ASR confirms '免税商品' (duty-free goods) correcting '棉市' (cotton market), matching Japanese source." } ], "language": "zh", "packet_ids": [ 57 ], "index": 148, "independent_ASR": { "text": "各位贵宾，我们现在结束免税商品的销售，感谢您的购买。", "chunks": [ { "text": "各位贵宾，我们现在结束免税商品的销售，感谢您的购买。", "start": 0.0, "end": 5.7, "chunk_index": 0, "language": "Chinese", "finish_reason": "eos", "truncated": false, "generated_tokens": 14, "max_new_tokens": 128 } ], "detected_language": "Chinese", "segments": [ { "start": 944.6, "end": 950.3000000000001, "text": "各位贵宾，我们现在结束免税商品的销售，感谢您的购买。" } ], "start": 944.6, "end": 950.3, "engine": "mlx-qwen3-asr", "language": "zh" }, "raw_word_timestamps": [ { "word": "各位", "start": 945.26, "end": 945.6800000000001, "probability": 0.94677734375 }, { "word": "贵", "start": 945.6800000000001, "end": 945.86, "probability": 0.50830078125 }, { "word": "民", "start": 945.86, "end": 945.98, "probability": 0.52880859375 }, { "word": "我们", "start": 945.98, "end": 946.22, "probability": 0.646484375 }, { "word": "现在", "start": 946.22, "end": 946.5, "probability": 0.9921875 }, { "word": "结", "start": 946.5, "end": 946.7, "probability": 0.98876953125 }, { "word": "束", "start": 946.7, "end": 946.94, "probability": 0.99951171875 }, { "word": "棉", "start": 946.94, "end": 947.0600000000001, "probability": 0.6728515625 }, { "word": "市", "start": 947.0600000000001, "end": 947.2, "probability": 0.1600341796875 }, { "word": "上面", "start": 947.2, "end": 947.44, "probability": 0.78662109375 }, { "word": "的", "start": 947.44, "end": 947.66, "probability": 0.9990234375 }, { "word": "销", "start": 947.66, "end": 947.84, "probability": 0.99462890625 }, { "word": "售", "start": 947.84, "end": 948.08, "probability": 1.0 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "各位贵宾我们现在结束免税商品的销售", "review_status": "CERTAIN_RESTORATION", "classification": "CERTAIN_RESTORATION", "confidence": 0.95, "correction_reason": "Independent ASR confirms '免税商品' (duty-free goods) correcting '棉市' (cotton market), matching Japanese source." }, { "start": 948.08, "end": 949.1800000000001, "text": "感谢你的购买", "raw_asr_text": "感谢你的购买", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 57, "words": [ { "word": "感", "start": 948.08, "end": 948.26, "probability": 0.78271484375 }, { "word": "谢", "start": 948.26, "end": 948.4200000000001, "probability": 0.998046875 }, { "word": "你的", "start": 948.4200000000001, "end": 948.64, "probability": 0.83935546875 }, { "word": "购", "start": 948.64, "end": 948.82, "probability": 0.9990234375 }, { "word": "买", "start": 948.82, "end": 949.1800000000001, "probability": 1.0 } ], "asr_metrics": { "avg_logprob": -0.36127067434376686, "compression_ratio": 0.8625, "no_speech_prob": 1.360169268882494e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "各位贵民我们现在结束棉市上面的销售感谢你的购买" } ], "id": 149, "review_annotation": "", "row_id": 149, "bounded_row_id": 149, "parent_row_ids": [ 149 ], "parent_row_id": 149, "packet_id": 57, "LID_evidence": [ { "region_id": 139, "probabilities": { "ja": 0.012560169550124557, "zh": 0.8904412806034088, "en": 0.028701536101289095 }, "status": "STABLE" } ], "VAD_region_ids": [ 139 ], "language_confidence": 0.8904412806034088, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 944.6, "packet_end": 950.3, "raw_text": "感谢你的购买", "selected_text": "感谢你的购买", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 149 ], "packet_id": 57, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "感谢你的购买", "after": "感谢你的购买", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard polite closing. Matches Japanese 'お買い上げありがとうございました'." } ], "language": "zh", "packet_ids": [ 57 ], "index": 149, "source_text": "感谢你的购买", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.9, "correction_reason": "Standard polite closing. Matches Japanese 'お買い上げありがとうございました'." }, { "start": 952.96, "end": 957.58, "text": "皆様、シートベルトは緩みのないようしっかりお締めください。", "raw_asr_text": "皆様、シートベルトは緩みのないようしっかりお締めください。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 58, "words": [ { "word": "皆", "start": 952.96, "end": 953.28, "probability": 0.8916015625 }, { "word": "様、", "start": 953.28, "end": 953.5400000000001, "probability": 0.6298828125 }, { "word": "シ", "start": 954.14, "end": 954.5, "probability": 0.99951171875 }, { "word": "ート", "start": 954.5, "end": 954.6800000000001, "probability": 1.0 }, { "word": "ベ", "start": 954.6800000000001, "end": 954.84, "probability": 0.99951171875 }, { "word": "ル", "start": 954.84, "end": 954.9200000000001, "probability": 0.99951171875 }, { "word": "ト", "start": 954.9200000000001, "end": 955.0400000000001, "probability": 0.99951171875 }, { "word": "は", "start": 955.0400000000001, "end": 955.22, "probability": 1.0 }, { "word": "緩", "start": 955.22, "end": 955.76, "probability": 0.9755859375 }, { "word": "み", "start": 955.76, "end": 955.9, "probability": 1.0 }, { "word": "の", "start": 955.9, "end": 956.02, "probability": 1.0 }, { "word": "ない", "start": 956.02, "end": 956.22, "probability": 0.99951171875 }, { "word": "よう", "start": 956.22, "end": 956.38, "probability": 0.99169921875 }, { "word": "し", "start": 956.38, "end": 956.58, "probability": 0.8544921875 }, { "word": "っ", "start": 956.58, "end": 956.6800000000001, "probability": 1.0 }, { "word": "か", "start": 956.6800000000001, "end": 956.78, "probability": 1.0 }, { "word": "り", "start": 956.78, "end": 956.88, "probability": 1.0 }, { "word": "お", "start": 956.88, "end": 957.0, "probability": 0.98828125 }, { "word": "締", "start": 957.0, "end": 957.16, "probability": 0.9931640625 }, { "word": "め", "start": 957.16, "end": 957.24, "probability": 1.0 }, { "word": "ください。", "start": 957.24, "end": 957.58, "probability": 0.994140625 } ], "asr_metrics": { "avg_logprob": -0.12805817808423722, "compression_ratio": 1.0481927710843373, "no_speech_prob": 4.514403174199488e-12 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "皆様、シートベルトは緩みのないようしっかりお締めください。" } ], "id": 150, "review_annotation": "", "row_id": 150, "bounded_row_id": 150, "parent_row_ids": [ 150 ], "parent_row_id": 150, "packet_id": 58, "LID_evidence": [ { "region_id": 140, "probabilities": { "ja": 0.9871276319026947, "zh": 0.0006956182623980567, "en": 0.007129889389034361 }, "status": "UNCERTAIN" }, { "region_id": 141, "probabilities": { "ja": 0.8961294862886213, "zh": 0.000697063135290095, "en": 0.09631357955301857 }, "status": "STABLE" } ], "VAD_region_ids": [ 140, 141 ], "language_confidence": 0.9871276319026947, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 952.1, "packet_end": 958.8, "raw_text": "皆様、シートベルトは緩みのないようしっかりお締めください。", "selected_text": "皆様、シートベルトは緩みのないようしっかりお締めください。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 150 ], "packet_id": 58, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "皆様、シートベルトは緩みのないようしっかりお締めください。", "after": "皆様、シートベルトは緩みのないようしっかりお締めください。", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Text matches standard ANA announcement phrasing and context perfectly." } ], "language": "ja", "packet_ids": [ 58 ], "index": 150, "source_text": "皆様、シートベルトは緩みのないようしっかりお締めください。", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Text matches standard ANA announcement phrasing and context perfectly." }, { "start": 958.9, "end": 963.5600000000001, "text": "Ladies and gentlemen, please make sure that your seat belt is securely fastened.", "raw_asr_text": "Ladies and gentlemen, please make sure that your seat belt is securely fastened.", "source_language": "en", "source_lang": "en", "asr_packet_id": 59, "words": [], "asr_metrics": { "avg_logprob": -0.33229517936706543, "compression_ratio": 1.0963855421686748, "no_speech_prob": 1.7402588701684318e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "en", "text": "Ladies and gentlemen, please make sure that your seat belt is securely fastened. Thank you." } ], "id": 151, "review_annotation": "", "row_id": 151, "bounded_row_id": 151, "parent_row_ids": [ 151 ], "parent_row_id": 151, "packet_id": 59, "LID_evidence": [ { "region_id": 142, "probabilities": { "ja": 0.1540860772583983, "zh": 0.0009699663157031561, "en": 0.833804544682304 }, "status": "STABLE" }, { "region_id": 143, "probabilities": { "ja": 0.0002474130186682386, "zh": 0.00013525649612549588, "en": 0.998000356886122 }, "status": "STABLE" } ], "VAD_region_ids": [ 142, 143 ], "language_confidence": 0.998000356886122, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 958.1, "packet_end": 965.5999999999999, "raw_text": "Ladies and gentlemen, please make sure that your seat belt is securely fastened.", "selected_text": "Ladies and gentlemen, please make sure that your seat belt is securely fastened.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 151 ], "packet_id": 59, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "Ladies and gentlemen, please make sure that your seat belt is securely fastened.", "after": "Ladies and gentlemen, please make sure that your seat belt is securely fastened.", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Matches standard English announcement and Chinese translation context." } ], "language": "en", "packet_ids": [ 59 ], "index": 151, "source_text": "Ladies and gentlemen, please make sure that your seat belt is securely fastened.", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Matches standard English announcement and Chinese translation context.", "raw_word_timestamps": [ { "word": " Ladies", "start": 958.72, "end": 959.26, "probability": 0.60546875 }, { "word": " and", "start": 959.26, "end": 959.52, "probability": 0.99462890625 }, { "word": " gentlemen,", "start": 959.52, "end": 959.9, "probability": 0.8623046875 }, { "word": " please", "start": 960.32, "end": 960.9200000000001, "probability": 0.9755859375 }, { "word": " make", "start": 960.9200000000001, "end": 961.14, "probability": 0.99755859375 }, { "word": " sure", "start": 961.14, "end": 961.46, "probability": 0.998046875 }, { "word": " that", "start": 961.46, "end": 961.7, "probability": 0.98388671875 }, { "word": " your", "start": 961.7, "end": 961.84, "probability": 0.994140625 }, { "word": " seat", "start": 961.84, "end": 962.08, "probability": 0.9970703125 }, { "word": " belt", "start": 962.08, "end": 962.28, "probability": 0.39697265625 }, { "word": " is", "start": 962.28, "end": 962.44, "probability": 0.99853515625 }, { "word": " securely", "start": 962.44, "end": 962.86, "probability": 0.99658203125 }, { "word": " fastened.", "start": 962.86, "end": 963.5600000000001, "probability": 0.99853515625 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 963.8000000000001, "end": 964.5, "text": "Thank you.", "raw_asr_text": "Thank you.", "source_language": "en", "source_lang": "en", "asr_packet_id": 59, "words": [ { "word": " Thank", "start": 963.8000000000001, "end": 964.22, "probability": 0.9990234375 }, { "word": " you.", "start": 964.22, "end": 964.5, "probability": 0.99755859375 } ], "asr_metrics": { "avg_logprob": -0.33229517936706543, "compression_ratio": 1.0963855421686748, "no_speech_prob": 1.7402588701684318e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "en", "text": "Ladies and gentlemen, please make sure that your seat belt is securely fastened. Thank you." } ], "id": 152, "review_annotation": "", "row_id": 152, "bounded_row_id": 152, "parent_row_ids": [ 152 ], "parent_row_id": 152, "packet_id": 59, "LID_evidence": [ { "region_id": 143, "probabilities": { "ja": 0.0002474130186682386, "zh": 0.00013525649612549588, "en": 0.998000356886122 }, "status": "STABLE" } ], "VAD_region_ids": [ 143 ], "language_confidence": 0.998000356886122, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 958.1, "packet_end": 965.5999999999999, "raw_text": "Thank you.", "selected_text": "Thank you.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 152 ], "packet_id": 59, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "Thank you.", "after": "Thank you.", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard closing phrase, supported by surrounding context." } ], "language": "en", "packet_ids": [ 59 ], "index": 152, "source_text": "Thank you.", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard closing phrase, supported by surrounding context." }, { "start": 971.0, "end": 975.8000000000001, "text": "皆様、ベルト着用サインが点灯いたしました。揺れることが予想されております。", "raw_asr_text": "皆様、ベルト着用サインが点灯いたしました。揺れることが予想されております。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 60, "words": [], "asr_metrics": { "avg_logprob": -0.10725033283233643, "compression_ratio": 1.3175675675675675, "no_speech_prob": 5.621176701192665e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "皆様、ベルト着用サインが点灯いたしました。揺れることが予想されております。ステートベルトは緩みのないよう、しっかりお締めください。" } ], "id": 153, "review_annotation": "", "row_id": 153, "bounded_row_id": 153, "parent_row_ids": [ 153 ], "parent_row_id": 153, "packet_id": 60, "LID_evidence": [ { "region_id": 144, "probabilities": { "ja": 0.9846072793006897, "zh": 0.0007587256753010609, "en": 0.008813211592496373 }, "status": "STABLE" } ], "VAD_region_ids": [ 144 ], "language_confidence": 0.9846072793006897, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 970.2, "packet_end": 980.4, "raw_text": "皆様、ベルト着用サインが点灯いたしました。揺れることが予想されております。", "selected_text": "皆様、ベルト着用サインが点灯いたしました。揺れることが予想されております。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 153 ], "packet_id": 60, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "皆様、ベルト着用サインが点灯いたしました。揺れることが予想されております。", "after": "皆様、ベルト着用サインが点灯いたしました。揺れることが予想されております。", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard Japanese announcement text, consistent with context." } ], "language": "ja", "packet_ids": [ 60 ], "index": 153, "source_text": "皆様、ベルト着用サインが点灯いたしました。揺れることが予想されております。", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard Japanese announcement text, consistent with context.", "raw_word_timestamps": [ { "word": "皆", "start": 970.96, "end": 971.24, "probability": 0.91162109375 }, { "word": "様、", "start": 971.24, "end": 971.5, "probability": 0.73388671875 }, { "word": "ベ", "start": 971.62, "end": 971.7, "probability": 0.99951171875 }, { "word": "ル", "start": 971.7, "end": 971.7800000000001, "probability": 1.0 }, { "word": "ト", "start": 971.7800000000001, "end": 971.88, "probability": 1.0 }, { "word": "着", "start": 971.88, "end": 972.0200000000001, "probability": 1.0 }, { "word": "用", "start": 972.0200000000001, "end": 972.24, "probability": 1.0 }, { "word": "サ", "start": 972.24, "end": 972.44, "probability": 0.99951171875 }, { "word": "イ", "start": 972.44, "end": 972.5200000000001, "probability": 1.0 }, { "word": "ン", "start": 972.5200000000001, "end": 972.5600000000001, "probability": 1.0 }, { "word": "が", "start": 972.5600000000001, "end": 972.6600000000001, "probability": 1.0 }, { "word": "点", "start": 972.6600000000001, "end": 972.82, "probability": 0.98681640625 }, { "word": "灯", "start": 972.82, "end": 973.0200000000001, "probability": 0.9970703125 }, { "word": "いた", "start": 973.0200000000001, "end": 973.1600000000001, "probability": 0.978515625 }, { "word": "しました。", "start": 973.1600000000001, "end": 973.58, "probability": 1.0 }, { "word": "揺", "start": 974.0, "end": 974.1, "probability": 0.9892578125 }, { "word": "れる", "start": 974.1, "end": 974.26, "probability": 1.0 }, { "word": "こと", "start": 974.26, "end": 974.46, "probability": 0.99951171875 }, { "word": "が", "start": 974.46, "end": 974.6800000000001, "probability": 1.0 }, { "word": "予", "start": 974.6800000000001, "end": 974.9200000000001, "probability": 1.0 }, { "word": "想", "start": 974.9200000000001, "end": 975.0200000000001, "probability": 1.0 }, { "word": "さ", "start": 975.0200000000001, "end": 975.1600000000001, "probability": 1.0 }, { "word": "れて", "start": 975.1600000000001, "end": 975.34, "probability": 1.0 }, { "word": "お", "start": 975.34, "end": 975.48, "probability": 0.9990234375 }, { "word": "ります。", "start": 975.48, "end": 975.8000000000001, "probability": 1.0 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 976.2, "end": 979.0400000000001, "text": "ステートベルトは緩みのないよう、しっかりお締めください。", "raw_asr_text": "ステートベルトは緩みのないよう、しっかりお締めください。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 60, "words": [ { "word": "ス", "start": 976.2, "end": 976.3000000000001, "probability": 0.8134765625 }, { "word": "テ", "start": 976.3000000000001, "end": 976.36, "probability": 0.415771484375 }, { "word": "ート", "start": 976.36, "end": 976.48, "probability": 0.77490234375 }, { "word": "ベ", "start": 976.48, "end": 976.6400000000001, "probability": 0.99951171875 }, { "word": "ル", "start": 976.6400000000001, "end": 976.7, "probability": 1.0 }, { "word": "ト", "start": 976.7, "end": 976.8000000000001, "probability": 1.0 }, { "word": "は", "start": 976.8000000000001, "end": 976.98, "probability": 0.99951171875 }, { "word": "緩", "start": 976.98, "end": 977.22, "probability": 0.783203125 }, { "word": "み", "start": 977.22, "end": 977.36, "probability": 0.99951171875 }, { "word": "の", "start": 977.36, "end": 977.48, "probability": 1.0 }, { "word": "ない", "start": 977.48, "end": 977.6800000000001, "probability": 0.9990234375 }, { "word": "よう、", "start": 977.6800000000001, "end": 977.84, "probability": 0.98828125 }, { "word": "し", "start": 978.0, "end": 978.0600000000001, "probability": 0.99853515625 }, { "word": "っ", "start": 978.0600000000001, "end": 978.1, "probability": 1.0 }, { "word": "か", "start": 978.1, "end": 978.1600000000001, "probability": 1.0 }, { "word": "り", "start": 978.1600000000001, "end": 978.26, "probability": 1.0 }, { "word": "お", "start": 978.26, "end": 978.44, "probability": 0.99169921875 }, { "word": "締", "start": 978.44, "end": 978.5600000000001, "probability": 0.9921875 }, { "word": "め", "start": 978.5600000000001, "end": 978.6600000000001, "probability": 0.99951171875 }, { "word": "ください。", "start": 978.6600000000001, "end": 979.0400000000001, "probability": 0.9970703125 } ], "asr_metrics": { "avg_logprob": -0.10725033283233643, "compression_ratio": 1.3175675675675675, "no_speech_prob": 5.621176701192665e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "皆様、ベルト着用サインが点灯いたしました。揺れることが予想されております。ステートベルトは緩みのないよう、しっかりお締めください。" } ], "id": 154, "review_annotation": "待確認", "row_id": 154, "bounded_row_id": 154, "parent_row_ids": [ 154 ], "parent_row_id": 154, "packet_id": 60, "LID_evidence": [ { "region_id": 144, "probabilities": { "ja": 0.9846072793006897, "zh": 0.0007587256753010609, "en": 0.008813211592496373 }, "status": "STABLE" } ], "VAD_region_ids": [ 144 ], "language_confidence": 0.9846072793006897, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 970.2, "packet_end": 980.4, "raw_text": "ステートベルトは緩みのないよう、しっかりお締めください。", "selected_text": "ステートベルトは緩みのないよう、しっかりお締めください。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 154 ], "packet_id": 60, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "ステートベルトは緩みのないよう、しっかりお締めください。", "after": "ステートベルトは緩みのないよう、しっかりお締めください。", "classification": "BEST_SUPPORTED_WITH_REVIEW", "reason": "Independent ASR says 'シェッドベルト' (shed belt); likely 'シートベルト' (seat belt). Marked for review." } ], "language": "ja", "packet_ids": [ 60 ], "index": 154, "independent_ASR": { "text": "皆様、ベルト着用サインが点灯いたしました。揺れることが予想されております。シェッドベルトは緩みのないようしっかりお締めください。", "chunks": [ { "text": "皆様、ベルト着用サインが点灯いたしました。揺れることが予想されております。シェッドベルトは緩みのないようしっかりお締めください。", "start": 0.0, "end": 10.2, "chunk_index": 0, "language": "Japanese", "finish_reason": "eos", "truncated": false, "generated_tokens": 41, "max_new_tokens": 155 } ], "detected_language": "Japanese", "segments": [ { "start": 970.2, "end": 980.4000000000001, "text": "皆様、ベルト着用サインが点灯いたしました。揺れることが予想されております。シェッドベルトは緩みのないようしっかりお締めください。" } ], "start": 970.2, "end": 980.4, "engine": "mlx-qwen3-asr", "language": "ja" }, "source_text": "ステートベルトは緩みのないよう、しっかりお締めください。", "review_status": "BEST_SUPPORTED_WITH_REVIEW", "classification": "BEST_SUPPORTED_WITH_REVIEW", "confidence": 0.8, "correction_reason": "Independent ASR says 'シェッドベルト' (shed belt); likely 'シートベルト' (seat belt). Marked for review." }, { "start": 982.4, "end": 985.78, "text": "Ladies and gentlemen, the captain has turned on the seatbelt sign.", "raw_asr_text": "Ladies and gentlemen, the captain has turned on the seatbelt sign.", "source_language": "en", "source_lang": "en", "asr_packet_id": 61, "words": [], "asr_metrics": { "avg_logprob": -0.2843520781573127, "compression_ratio": 1.0, "no_speech_prob": 2.093694488614073e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "en", "text": "Ladies and gentlemen, the captain has turned on the seatbelt sign." } ], "id": 155, "review_annotation": "", "row_id": 155, "bounded_row_id": 155, "parent_row_ids": [ 155 ], "parent_row_id": 155, "packet_id": 61, "LID_evidence": [ { "region_id": 145, "probabilities": { "ja": 0.003964315017219633, "zh": 0.0019905566703528168, "en": 0.9690362691879273 }, "status": "STABLE" } ], "VAD_region_ids": [ 145 ], "language_confidence": 0.9690362691879273, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 981.6, "packet_end": 987.0, "raw_text": "Ladies and gentlemen, the captain has turned on the seatbelt sign.", "selected_text": "Ladies and gentlemen, the captain has turned on the seatbelt sign.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 155 ], "packet_id": 61, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "Ladies and gentlemen, the captain has turned on the seatbelt sign.", "after": "Ladies and gentlemen, the captain has turned on the seatbelt sign.", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Matches standard English announcement and Japanese context." } ], "language": "en", "packet_ids": [ 61 ], "index": 155, "source_text": "Ladies and gentlemen, the captain has turned on the seatbelt sign.", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Matches standard English announcement and Japanese context.", "raw_word_timestamps": [ { "word": " Ladies", "start": 982.24, "end": 982.76, "probability": 0.75 }, { "word": " and", "start": 982.76, "end": 983.08, "probability": 0.9931640625 }, { "word": " gentlemen,", "start": 983.08, "end": 983.44, "probability": 0.7998046875 }, { "word": " the", "start": 983.66, "end": 983.74, "probability": 0.9736328125 }, { "word": " captain", "start": 983.74, "end": 983.98, "probability": 0.8564453125 }, { "word": " has", "start": 983.98, "end": 984.38, "probability": 0.98583984375 }, { "word": " turned", "start": 984.38, "end": 984.64, "probability": 0.99755859375 }, { "word": " on", "start": 984.64, "end": 984.84, "probability": 0.9912109375 }, { "word": " the", "start": 984.84, "end": 984.98, "probability": 0.95849609375 }, { "word": " seatbelt", "start": 984.98, "end": 985.4, "probability": 0.92529296875 }, { "word": " sign.", "start": 985.4, "end": 985.78, "probability": 0.9892578125 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 988.3, "end": 993.0, "text": "Please return to your seat and make sure your seat belt is securely fastened. Thank you.", "raw_asr_text": "Please return to your seat and make sure your seat belt is securely fastened. Thank you.", "source_language": "en", "source_lang": "en", "asr_packet_id": 62, "words": [], "asr_metrics": { "avg_logprob": -0.3147157755765048, "compression_ratio": 1.1428571428571428, "no_speech_prob": 7.119967115887249e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "en", "text": "Please return to your seat and make sure your seat belt is securely fastened. Thank you." } ], "id": 156, "review_annotation": "", "row_id": 156, "bounded_row_id": 156, "parent_row_ids": [ 156 ], "parent_row_id": 156, "packet_id": 62, "LID_evidence": [ { "region_id": 146, "probabilities": { "ja": 0.042166848899796605, "zh": 0.07096997650594196, "en": 0.7485834563320334 }, "status": "STABLE" } ], "VAD_region_ids": [ 146 ], "language_confidence": 0.7485834563320334, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 987.5, "packet_end": 994.1999999999999, "raw_text": "Please return to your seat and make sure your seat belt is securely fastened. Thank you.", "selected_text": "Please return to your seat and make sure your seat belt is securely fastened. Thank you.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 156 ], "packet_id": 62, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "Please return to your seat and make sure your seat belt is securely fastened. Thank you.", "after": "Please return to your seat and make sure your seat belt is securely fastened. Thank you.", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Matches standard English announcement and Chinese translation context." } ], "language": "en", "packet_ids": [ 62 ], "index": 156, "source_text": "Please return to your seat and make sure your seat belt is securely fastened. Thank you.", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Matches standard English announcement and Chinese translation context.", "raw_word_timestamps": [ { "word": " Please", "start": 988.08, "end": 988.64, "probability": 0.72314453125 }, { "word": " return", "start": 988.64, "end": 988.96, "probability": 0.91650390625 }, { "word": " to", "start": 988.96, "end": 989.16, "probability": 0.99267578125 }, { "word": " your", "start": 989.16, "end": 989.36, "probability": 0.990234375 }, { "word": " seat", "start": 989.36, "end": 989.66, "probability": 0.9814453125 }, { "word": " and", "start": 989.66, "end": 989.84, "probability": 0.92626953125 }, { "word": " make", "start": 989.84, "end": 990.06, "probability": 0.99658203125 }, { "word": " sure", "start": 990.06, "end": 990.34, "probability": 0.99462890625 }, { "word": " your", "start": 990.34, "end": 990.6, "probability": 0.9697265625 }, { "word": " seat", "start": 990.6, "end": 990.9, "probability": 0.99658203125 }, { "word": " belt", "start": 990.9, "end": 991.22, "probability": 0.6279296875 }, { "word": " is", "start": 991.22, "end": 991.4, "probability": 0.99658203125 }, { "word": " securely", "start": 991.4, "end": 991.72, "probability": 0.8095703125 }, { "word": " fastened.", "start": 991.72, "end": 992.4, "probability": 0.93408203125 }, { "word": " Thank", "start": 992.58, "end": 992.72, "probability": 0.98583984375 }, { "word": " you.", "start": 992.72, "end": 993.0, "probability": 0.97509765625 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 993.9, "end": 996.64, "text": "各位貴賓，現在安全帶指示燈已經亮起，請您確實系好安全帶，並且請勿使用洗手間，謝謝。", "raw_asr_text": "各位队宾,现在安全带这扇灯已经亮起", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 63, "words": [], "asr_metrics": { "avg_logprob": -0.2518552541732788, "compression_ratio": 0.9734513274336283, "no_speech_prob": 1.2286080139367606e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "各位队宾,现在安全带这扇灯已经亮起请您确实系好安全带并且请勿使用洗手间,谢谢" } ], "id": 157, "review_annotation": "", "row_id": 157, "bounded_row_id": 157, "parent_row_ids": [ 157 ], "parent_row_id": 157, "packet_id": 63, "LID_evidence": [ { "region_id": 147, "probabilities": { "ja": 0.030756894157578547, "zh": 0.8363180972635746, "en": 0.045526009619546434 }, "status": "STABLE" } ], "VAD_region_ids": [ 147 ], "language_confidence": 0.8363180972635746, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 993.1, "packet_end": 1002.1999999999999, "raw_text": "各位队宾,现在安全带这扇灯已经亮起", "selected_text": "各位貴賓，現在安全帶指示燈已經亮起，請您確實系好安全帶，並且請勿使用洗手間，謝謝。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 157 ], "packet_id": 63, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "各位队宾,现在安全带这扇灯已经亮起", "after": "各位貴賓，現在安全帶指示燈已經亮起，請您確實系好安全帶，並且請勿使用洗手間，謝謝。", "classification": "CERTAIN_RESTORATION", "reason": "Independent ASR and context confirm 'indicator light' and 'do not use lavatory' despite source errors." } ], "language": "zh", "packet_ids": [ 63 ], "index": 157, "independent_ASR": { "text": "各位贵宾，现在安全带指示灯已经亮起，请您确实系好安全带，并且请勿使用洗手间，谢谢。", "chunks": [ { "text": "各位贵宾，现在安全带指示灯已经亮起，请您确实系好安全带，并且请勿使用洗手间，谢谢。", "start": 0.0, "end": 9.1, "chunk_index": 0, "language": "Chinese", "finish_reason": "eos", "truncated": false, "generated_tokens": 28, "max_new_tokens": 142 } ], "detected_language": "Chinese", "segments": [ { "start": 993.1, "end": 1002.2, "text": "各位贵宾，现在安全带指示灯已经亮起，请您确实系好安全带，并且请勿使用洗手间，谢谢。" } ], "start": 993.1, "end": 1002.1999999999999, "engine": "mlx-qwen3-asr", "language": "zh" }, "raw_word_timestamps": [ { "word": "各位", "start": 993.8000000000001, "end": 994.2, "probability": 0.94775390625 }, { "word": "队", "start": 994.2, "end": 994.38, "probability": 0.5703125 }, { "word": "宾,", "start": 994.38, "end": 994.5400000000001, "probability": 0.587890625 }, { "word": "现在", "start": 994.6800000000001, "end": 994.86, "probability": 0.9638671875 }, { "word": "安全", "start": 994.86, "end": 995.24, "probability": 0.974609375 }, { "word": "带", "start": 995.24, "end": 995.48, "probability": 0.99072265625 }, { "word": "这", "start": 995.48, "end": 995.64, "probability": 0.1856689453125 }, { "word": "扇", "start": 995.64, "end": 995.74, "probability": 0.376708984375 }, { "word": "灯", "start": 995.74, "end": 995.9, "probability": 0.9306640625 }, { "word": "已经", "start": 995.9, "end": 996.1, "probability": 0.9970703125 }, { "word": "亮", "start": 996.1, "end": 996.34, "probability": 0.99462890625 }, { "word": "起", "start": 996.34, "end": 996.64, "probability": 0.9970703125 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "各位貴賓，現在安全帶指示燈已經亮起，請您確實系好安全帶，並且請勿使用洗手間，謝謝。", "review_status": "CERTAIN_RESTORATION", "classification": "CERTAIN_RESTORATION", "confidence": 0.95, "correction_reason": "Independent ASR and context confirm 'indicator light' and 'do not use lavatory' despite source errors." }, { "start": 996.64, "end": 998.84, "text": "请您确实系好安全带", "raw_asr_text": "请您确实系好安全带", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 63, "words": [], "asr_metrics": { "avg_logprob": -0.2518552541732788, "compression_ratio": 0.9734513274336283, "no_speech_prob": 1.2286080139367606e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "各位队宾,现在安全带这扇灯已经亮起请您确实系好安全带并且请勿使用洗手间,谢谢" } ], "id": 158, "review_annotation": "", "row_id": 158, "bounded_row_id": 158, "parent_row_ids": [ 158 ], "parent_row_id": 158, "packet_id": 63, "LID_evidence": [ { "region_id": 147, "probabilities": { "ja": 0.030756894157578547, "zh": 0.8363180972635746, "en": 0.045526009619546434 }, "status": "STABLE" } ], "VAD_region_ids": [ 147 ], "language_confidence": 0.8363180972635746, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 993.1, "packet_end": 1002.1999999999999, "raw_text": "请您确实系好安全带", "selected_text": "请您确实系好安全带", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 158 ], "packet_id": 63, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "请您确实系好安全带", "after": "请您确实系好安全带", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard Chinese announcement phrasing, supported by context." } ], "language": "zh", "packet_ids": [ 63 ], "index": 158, "source_text": "请您确实系好安全带", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard Chinese announcement phrasing, supported by context.", "raw_word_timestamps": [ { "word": "请", "start": 996.64, "end": 997.16, "probability": 0.63037109375 }, { "word": "您", "start": 997.16, "end": 997.32, "probability": 0.9951171875 }, { "word": "确", "start": 997.32, "end": 997.52, "probability": 0.9990234375 }, { "word": "实", "start": 997.52, "end": 997.7, "probability": 0.9990234375 }, { "word": "系", "start": 997.7, "end": 997.7, "probability": 0.98681640625 }, { "word": "好", "start": 997.7, "end": 998.0400000000001, "probability": 0.99267578125 }, { "word": "安全", "start": 998.0400000000001, "end": 998.36, "probability": 0.998046875 }, { "word": "带", "start": 998.36, "end": 998.84, "probability": 0.9951171875 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 998.84, "end": 1000.88, "text": "并且请勿使用洗手间,谢谢", "raw_asr_text": "并且请勿使用洗手间,谢谢", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 63, "words": [ { "word": "并", "start": 998.84, "end": 999.16, "probability": 0.837890625 }, { "word": "且", "start": 999.16, "end": 999.3000000000001, "probability": 0.99951171875 }, { "word": "请", "start": 999.3000000000001, "end": 999.48, "probability": 0.89501953125 }, { "word": "勿", "start": 999.48, "end": 999.64, "probability": 0.98046875 }, { "word": "使", "start": 999.64, "end": 999.8000000000001, "probability": 0.9619140625 }, { "word": "用", "start": 999.8000000000001, "end": 1000.02, "probability": 1.0 }, { "word": "洗", "start": 1000.02, "end": 1000.12, "probability": 0.9990234375 }, { "word": "手", "start": 1000.12, "end": 1000.32, "probability": 1.0 }, { "word": "间,", "start": 1000.32, "end": 1000.52, "probability": 0.986328125 }, { "word": "谢谢", "start": 1000.64, "end": 1000.88, "probability": 0.99755859375 } ], "asr_metrics": { "avg_logprob": -0.2518552541732788, "compression_ratio": 0.9734513274336283, "no_speech_prob": 1.2286080139367606e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "各位队宾,现在安全带这扇灯已经亮起请您确实系好安全带并且请勿使用洗手间,谢谢" } ], "id": 159, "review_annotation": "", "row_id": 159, "bounded_row_id": 159, "parent_row_ids": [ 159 ], "parent_row_id": 159, "packet_id": 63, "LID_evidence": [ { "region_id": 147, "probabilities": { "ja": 0.030756894157578547, "zh": 0.8363180972635746, "en": 0.045526009619546434 }, "status": "STABLE" } ], "VAD_region_ids": [ 147 ], "language_confidence": 0.8363180972635746, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 993.1, "packet_end": 1002.1999999999999, "raw_text": "并且请勿使用洗手间,谢谢", "selected_text": "并且请勿使用洗手间,谢谢", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 159 ], "packet_id": 63, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "并且请勿使用洗手间,谢谢", "after": "并且请勿使用洗手间,谢谢", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard Chinese announcement phrasing, supported by context." } ], "language": "zh", "packet_ids": [ 63 ], "index": 159, "source_text": "并且请勿使用洗手间,谢谢", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard Chinese announcement phrasing, supported by context." }, { "start": 1003.94, "end": 1006.64, "text": "皆様、ベルト着用サインは消えましたが、", "raw_asr_text": "皆様、ベルト着用サインは消えましたが、", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 64, "words": [ { "word": "皆", "start": 1003.94, "end": 1004.26, "probability": 0.90185546875 }, { "word": "様、", "start": 1004.26, "end": 1004.52, "probability": 0.69580078125 }, { "word": "ベ", "start": 1004.6, "end": 1004.76, "probability": 0.9990234375 }, { "word": "ル", "start": 1004.76, "end": 1004.82, "probability": 1.0 }, { "word": "ト", "start": 1004.82, "end": 1004.9200000000001, "probability": 0.99951171875 }, { "word": "着", "start": 1004.9200000000001, "end": 1005.08, "probability": 0.99951171875 }, { "word": "用", "start": 1005.08, "end": 1005.34, "probability": 0.99951171875 }, { "word": "サ", "start": 1005.34, "end": 1005.52, "probability": 0.8291015625 }, { "word": "イ", "start": 1005.52, "end": 1005.6, "probability": 0.99951171875 }, { "word": "ン", "start": 1005.6, "end": 1005.7, "probability": 0.99609375 }, { "word": "は", "start": 1005.7, "end": 1005.76, "probability": 0.994140625 }, { "word": "消", "start": 1005.76, "end": 1005.94, "probability": 0.9931640625 }, { "word": "え", "start": 1005.94, "end": 1006.1, "probability": 1.0 }, { "word": "ました", "start": 1006.1, "end": 1006.4200000000001, "probability": 1.0 }, { "word": "が、", "start": 1006.4200000000001, "end": 1006.64, "probability": 0.998046875 } ], "asr_metrics": { "avg_logprob": -0.22793259620666503, "compression_ratio": 0.8636363636363636, "no_speech_prob": 4.3423546008947156e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "皆様、ベルト着用サインは消えましたが、" } ], "id": 160, "review_annotation": "", "row_id": 160, "bounded_row_id": 160, "parent_row_ids": [ 160 ], "parent_row_id": 160, "packet_id": 64, "LID_evidence": [ { "region_id": 148, "probabilities": { "ja": 0.997534434000651, "zh": 0.00017649193129424626, "en": 0.0011514985429433484 }, "status": "STABLE" } ], "VAD_region_ids": [ 148 ], "language_confidence": 0.997534434000651, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1003.1, "packet_end": 1007.9, "raw_text": "皆様、ベルト着用サインは消えましたが、", "selected_text": "皆様、ベルト着用サインは消えましたが、", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 160 ], "packet_id": 64, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "皆様、ベルト着用サインは消えましたが、", "after": "皆様、ベルト着用サインは消えましたが、", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Matches standard Japanese announcement phrasing and context perfectly." } ], "language": "ja", "packet_ids": [ 64 ], "index": 160, "source_text": "皆様、ベルト着用サインは消えましたが、", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Matches standard Japanese announcement phrasing and context perfectly." }, { "start": 1009.8, "end": 1012.2, "text": "まもなく着陸体制に入ります。", "raw_asr_text": "まもなく着陸体制に入ります。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 65, "words": [], "asr_metrics": { "avg_logprob": -0.13397013346354167, "compression_ratio": 1.1596638655462186, "no_speech_prob": 5.301336626628572e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "まもなく着陸体制に入ります。ご用のないお客様は、シートベルトをお締めのままお過ごしください。" } ], "id": 161, "review_annotation": "", "row_id": 161, "bounded_row_id": 161, "parent_row_ids": [ 161 ], "parent_row_id": 161, "packet_id": 65, "LID_evidence": [ { "region_id": 149, "probabilities": { "ja": 0.7674929946660995, "zh": 0.03294918980585256, "en": 0.03815776909080645 }, "status": "STABLE" } ], "VAD_region_ids": [ 149 ], "language_confidence": 0.7674929946660995, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1009.0, "packet_end": 1019.1999999999999, "raw_text": "まもなく着陸体制に入ります。", "selected_text": "まもなく着陸体制に入ります。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 161 ], "packet_id": 65, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "まもなく着陸体制に入ります。", "after": "まもなく着陸体制に入ります。", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard aviation terminology confirmed by context and parallel text." } ], "language": "ja", "packet_ids": [ 65 ], "index": 161, "source_text": "まもなく着陸体制に入ります。", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard aviation terminology confirmed by context and parallel text.", "raw_word_timestamps": [ { "word": "ま", "start": 1009.66, "end": 1009.98, "probability": 0.88720703125 }, { "word": "も", "start": 1009.98, "end": 1010.12, "probability": 0.9990234375 }, { "word": "なく", "start": 1010.12, "end": 1010.36, "probability": 1.0 }, { "word": "着", "start": 1010.36, "end": 1010.6, "probability": 0.98876953125 }, { "word": "陸", "start": 1010.6, "end": 1010.9, "probability": 0.99951171875 }, { "word": "体", "start": 1010.9, "end": 1011.1, "probability": 0.82958984375 }, { "word": "制", "start": 1011.1, "end": 1011.32, "probability": 0.99560546875 }, { "word": "に", "start": 1011.32, "end": 1011.5, "probability": 0.99951171875 }, { "word": "入", "start": 1011.5, "end": 1011.72, "probability": 1.0 }, { "word": "ります。", "start": 1011.72, "end": 1012.2, "probability": 1.0 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 1013.62, "end": 1017.82, "text": "ご用のないお客様は、シートベルトをお締めのままお過ごしください。", "raw_asr_text": "ご用のないお客様は、シートベルトをお締めのままお過ごしください。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 65, "words": [ { "word": "ご", "start": 1013.62, "end": 1013.94, "probability": 0.73291015625 }, { "word": "用", "start": 1013.94, "end": 1014.06, "probability": 0.98974609375 }, { "word": "の", "start": 1014.06, "end": 1014.2, "probability": 0.9951171875 }, { "word": "ない", "start": 1014.2, "end": 1014.4, "probability": 0.99951171875 }, { "word": "お", "start": 1014.4, "end": 1014.56, "probability": 1.0 }, { "word": "客", "start": 1014.56, "end": 1014.68, "probability": 1.0 }, { "word": "様", "start": 1014.68, "end": 1014.98, "probability": 0.98974609375 }, { "word": "は、", "start": 1014.98, "end": 1015.26, "probability": 0.9990234375 }, { "word": "シ", "start": 1015.42, "end": 1015.5, "probability": 0.951171875 }, { "word": "ート", "start": 1015.5, "end": 1015.64, "probability": 1.0 }, { "word": "ベ", "start": 1015.64, "end": 1015.82, "probability": 0.9990234375 }, { "word": "ル", "start": 1015.82, "end": 1015.92, "probability": 0.99951171875 }, { "word": "ト", "start": 1015.92, "end": 1016.06, "probability": 1.0 }, { "word": "を", "start": 1016.06, "end": 1016.18, "probability": 0.99951171875 }, { "word": "お", "start": 1016.18, "end": 1016.32, "probability": 0.99560546875 }, { "word": "締", "start": 1016.32, "end": 1016.46, "probability": 0.990234375 }, { "word": "め", "start": 1016.46, "end": 1016.54, "probability": 1.0 }, { "word": "の", "start": 1016.54, "end": 1016.64, "probability": 1.0 }, { "word": "ま", "start": 1016.64, "end": 1016.8, "probability": 1.0 }, { "word": "ま", "start": 1016.8, "end": 1016.94, "probability": 1.0 }, { "word": "お", "start": 1016.94, "end": 1017.08, "probability": 0.99755859375 }, { "word": "過", "start": 1017.08, "end": 1017.24, "probability": 0.99951171875 }, { "word": "ご", "start": 1017.24, "end": 1017.32, "probability": 0.9990234375 }, { "word": "し", "start": 1017.32, "end": 1017.46, "probability": 0.998046875 }, { "word": "ください。", "start": 1017.46, "end": 1017.82, "probability": 1.0 } ], "asr_metrics": { "avg_logprob": -0.13397013346354167, "compression_ratio": 1.1596638655462186, "no_speech_prob": 5.301336626628572e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "まもなく着陸体制に入ります。ご用のないお客様は、シートベルトをお締めのままお過ごしください。" } ], "id": 162, "review_annotation": "", "row_id": 162, "bounded_row_id": 162, "parent_row_ids": [ 162 ], "parent_row_id": 162, "packet_id": 65, "LID_evidence": [ { "region_id": 150, "probabilities": { "ja": 0.790363609790802, "zh": 0.02603469534951728, "en": 0.030056732677621767 }, "status": "STABLE" } ], "VAD_region_ids": [ 150 ], "language_confidence": 0.790363609790802, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1009.0, "packet_end": 1019.1999999999999, "raw_text": "ご用のないお客様は、シートベルトをお締めのままお過ごしください。", "selected_text": "ご用のないお客様は、シートベルトをお締めのままお過ごしください。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 162 ], "packet_id": 65, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "ご用のないお客様は、シートベルトをお締めのままお過ごしください。", "after": "ご用のないお客様は、シートベルトをお締めのままお過ごしください。", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard phraseology for seatbelt instructions during descent." } ], "language": "ja", "packet_ids": [ 65 ], "index": 162, "source_text": "ご用のないお客様は、シートベルトをお締めのままお過ごしください。", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard phraseology for seatbelt instructions during descent." }, { "start": 1021.8, "end": 1027.54, "text": "化粧室ご使用ご希望のお客様は、ベルト着用サインが消えている間にご使用ください。", "raw_asr_text": "化粧室ご使用ご希望のお客様は、ベルト着用サインが消えている間にご使用ください。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 66, "words": [], "asr_metrics": { "avg_logprob": -0.12281765435871325, "compression_ratio": 1.0636363636363637, "no_speech_prob": 4.845234023026501e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "化粧室ご使用ご希望のお客様は、ベルト着用サインが消えている間にご使用ください。" } ], "id": 163, "review_annotation": "", "row_id": 163, "bounded_row_id": 163, "parent_row_ids": [ 163 ], "parent_row_id": 163, "packet_id": 66, "LID_evidence": [ { "region_id": 151, "probabilities": { "ja": 0.991778028011322, "zh": 0.0005376699147745967, "en": 0.004468670673668384 }, "status": "STABLE" } ], "VAD_region_ids": [ 151 ], "language_confidence": 0.991778028011322, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1021.0, "packet_end": 1028.8999999999999, "raw_text": "化粧室ご使用ご希望のお客様は、ベルト着用サインが消えている間にご使用ください。", "selected_text": "化粧室ご使用ご希望のお客様は、ベルト着用サインが消えている間にご使用ください。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 163 ], "packet_id": 66, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "化粧室ご使用ご希望のお客様は、ベルト着用サインが消えている間にご使用ください。", "after": "化粧室ご使用ご希望のお客様は、ベルト着用サインが消えている間にご使用ください。", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard instruction for lavatory use during descent phase." } ], "language": "ja", "packet_ids": [ 66 ], "index": 163, "source_text": "化粧室ご使用ご希望のお客様は、ベルト着用サインが消えている間にご使用ください。", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard instruction for lavatory use during descent phase.", "raw_word_timestamps": [ { "word": "化", "start": 1021.7, "end": 1022.06, "probability": 0.970703125 }, { "word": "粧", "start": 1022.06, "end": 1022.24, "probability": 0.9990234375 }, { "word": "室", "start": 1022.24, "end": 1022.5, "probability": 0.943359375 }, { "word": "ご", "start": 1022.5, "end": 1022.7, "probability": 0.8310546875 }, { "word": "使", "start": 1022.7, "end": 1022.9, "probability": 0.9736328125 }, { "word": "用", "start": 1022.9, "end": 1023.08, "probability": 1.0 }, { "word": "ご", "start": 1023.08, "end": 1023.24, "probability": 0.5205078125 }, { "word": "希望", "start": 1023.24, "end": 1023.54, "probability": 0.99951171875 }, { "word": "の", "start": 1023.54, "end": 1023.7, "probability": 1.0 }, { "word": "お", "start": 1023.7, "end": 1023.82, "probability": 1.0 }, { "word": "客", "start": 1023.82, "end": 1023.94, "probability": 1.0 }, { "word": "様", "start": 1023.94, "end": 1024.22, "probability": 0.994140625 }, { "word": "は、", "start": 1024.22, "end": 1024.42, "probability": 0.99560546875 }, { "word": "ベ", "start": 1024.5, "end": 1024.58, "probability": 0.9619140625 }, { "word": "ル", "start": 1024.58, "end": 1024.66, "probability": 0.99951171875 }, { "word": "ト", "start": 1024.66, "end": 1024.76, "probability": 1.0 }, { "word": "着", "start": 1024.76, "end": 1024.94, "probability": 1.0 }, { "word": "用", "start": 1024.94, "end": 1025.2, "probability": 0.99951171875 }, { "word": "サ", "start": 1025.2, "end": 1025.38, "probability": 0.9501953125 }, { "word": "イ", "start": 1025.38, "end": 1025.46, "probability": 1.0 }, { "word": "ン", "start": 1025.46, "end": 1025.52, "probability": 0.9990234375 }, { "word": "が", "start": 1025.52, "end": 1025.6, "probability": 1.0 }, { "word": "消", "start": 1025.6, "end": 1025.76, "probability": 0.99755859375 }, { "word": "えて", "start": 1025.76, "end": 1026.0, "probability": 1.0 }, { "word": "いる", "start": 1026.0, "end": 1026.2, "probability": 1.0 }, { "word": "間", "start": 1026.2, "end": 1026.34, "probability": 1.0 }, { "word": "に", "start": 1026.34, "end": 1026.66, "probability": 1.0 }, { "word": "ご", "start": 1026.66, "end": 1026.84, "probability": 0.98828125 }, { "word": "使", "start": 1026.84, "end": 1027.0, "probability": 0.99951171875 }, { "word": "用", "start": 1027.0, "end": 1027.12, "probability": 1.0 }, { "word": "ください。", "start": 1027.12, "end": 1027.54, "probability": 1.0 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 1033.2, "end": 1040.74, "text": "Ladies and gentlemen, the captain has turned off the seatbelt sign. However, please keep your seatbelt fastened whenever you're seated. Thank you.", "raw_asr_text": "Ladies and gentlemen, the captain has turned off the seatbelt sign. However, please keep your seatbelt fastened whenever you're seated. Thank you.", "source_language": "en", "source_lang": "en", "asr_packet_id": 67, "words": [], "asr_metrics": { "avg_logprob": -0.26244690838982077, "compression_ratio": 1.280701754385965, "no_speech_prob": 6.698599314125531e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "en", "text": "Ladies and gentlemen, the captain has turned off the seatbelt sign. However, please keep your seatbelt fastened whenever you're seated. Thank you." } ], "id": 164, "review_annotation": "", "row_id": 164, "bounded_row_id": 164, "parent_row_ids": [ 164 ], "parent_row_id": 164, "packet_id": 67, "LID_evidence": [ { "region_id": 152, "probabilities": { "ja": 0.011088373309646089, "zh": 0.0049035160873944945, "en": 0.8787001279684213 }, "status": "STABLE" } ], "VAD_region_ids": [ 152 ], "language_confidence": 0.8787001279684213, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1032.4, "packet_end": 1042.0, "raw_text": "Ladies and gentlemen, the captain has turned off the seatbelt sign. However, please keep your seatbelt fastened whenever you're seated. Thank you.", "selected_text": "Ladies and gentlemen, the captain has turned off the seatbelt sign. However, please keep your seatbelt fastened whenever you're seated. Thank you.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 164 ], "packet_id": 67, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "Ladies and gentlemen, the captain has turned off the seatbelt sign. However, please keep your seatbelt fastened whenever you're seated. Thank you.", "after": "Ladies and gentlemen, the captain has turned off the seatbelt sign. However, please keep your seatbelt fastened whenever you're seated. Thank you.", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard English announcement text matches context and grammar." } ], "language": "en", "packet_ids": [ 67 ], "index": 164, "source_text": "Ladies and gentlemen, the captain has turned off the seatbelt sign. However, please keep your seatbelt fastened whenever you're seated. Thank you.", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard English announcement text matches context and grammar.", "raw_word_timestamps": [ { "word": " Ladies", "start": 1032.4, "end": 1033.5600000000002, "probability": 0.70654296875 }, { "word": " and", "start": 1033.5600000000002, "end": 1033.88, "probability": 0.994140625 }, { "word": " gentlemen,", "start": 1033.88, "end": 1034.16, "probability": 0.84423828125 }, { "word": " the", "start": 1034.46, "end": 1034.46, "probability": 0.9794921875 }, { "word": " captain", "start": 1034.46, "end": 1034.7, "probability": 0.8857421875 }, { "word": " has", "start": 1034.7, "end": 1035.0400000000002, "probability": 0.9833984375 }, { "word": " turned", "start": 1035.0400000000002, "end": 1035.46, "probability": 0.9970703125 }, { "word": " off", "start": 1035.46, "end": 1035.8000000000002, "probability": 0.98583984375 }, { "word": " the", "start": 1035.8000000000002, "end": 1035.94, "probability": 0.99462890625 }, { "word": " seatbelt", "start": 1035.94, "end": 1036.38, "probability": 0.96923828125 }, { "word": " sign.", "start": 1036.38, "end": 1036.72, "probability": 0.60107421875 }, { "word": " However,", "start": 1036.92, "end": 1037.48, "probability": 0.9921875 }, { "word": " please", "start": 1037.76, "end": 1037.92, "probability": 0.99755859375 }, { "word": " keep", "start": 1037.92, "end": 1038.2, "probability": 0.97509765625 }, { "word": " your", "start": 1038.2, "end": 1038.38, "probability": 0.97509765625 }, { "word": " seatbelt", "start": 1038.38, "end": 1038.8200000000002, "probability": 0.994140625 }, { "word": " fastened", "start": 1038.8200000000002, "end": 1039.16, "probability": 0.70654296875 }, { "word": " whenever", "start": 1039.16, "end": 1039.42, "probability": 0.97998046875 }, { "word": " you're", "start": 1039.42, "end": 1039.7800000000002, "probability": 0.810546875 }, { "word": " seated.", "start": 1039.7800000000002, "end": 1040.0600000000002, "probability": 0.99951171875 }, { "word": " Thank", "start": 1040.3600000000001, "end": 1040.5, "probability": 0.9990234375 }, { "word": " you.", "start": 1040.5, "end": 1040.74, "probability": 0.9970703125 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 1045.9, "end": 1049.22, "text": "各位貴賓，雖然現在安全帶指示燈已經熄滅，但是由於飛行中可能突然遭遇亂流，建議在著座時仍確實系好安全帶。", "raw_asr_text": "各位貴賓,雖然現在安全帶只是在引進前面", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 68, "words": [], "asr_metrics": { "avg_logprob": -0.28191588906680837, "compression_ratio": 1.1306532663316582, "no_speech_prob": 3.142748214446378e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "各位貴賓,雖然現在安全帶只是在引進前面但是由於飛行中可能突然遭遇浪流簡易在救助時,人確實系好安全帶之後我們將準備降落如需要使用洗手間的貴賓請盡快使用,謝謝" } ], "id": 165, "review_annotation": "", "row_id": 165, "bounded_row_id": 165, "parent_row_ids": [ 165 ], "parent_row_id": 165, "packet_id": 68, "LID_evidence": [ { "region_id": 153, "probabilities": { "ja": 0.005180932303321444, "zh": 0.9626176589065127, "en": 0.023427424383246236 }, "status": "STABLE" } ], "VAD_region_ids": [ 153 ], "language_confidence": 0.9626176589065127, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1045.1000000000001, "packet_end": 1060.8999999999999, "raw_text": "各位貴賓,雖然現在安全帶只是在引進前面", "selected_text": "各位貴賓，雖然現在安全帶指示燈已經熄滅，但是由於飛行中可能突然遭遇亂流，建議在著座時仍確實系好安全帶。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 165 ], "packet_id": 68, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "各位貴賓,雖然現在安全帶只是在引進前面", "after": "各位貴賓，雖然現在安全帶指示燈已經熄滅，但是由於飛行中可能突然遭遇亂流，建議在著座時仍確實系好安全帶。", "classification": "CERTAIN_RESTORATION", "reason": "Independent ASR clarifies 'extinguished' and 'turbulence' replacing nonsensical source phrasing." } ], "language": "zh", "packet_ids": [ 68 ], "index": 165, "independent_ASR": { "text": "各位贵宾，虽然现在安全带指示灯已经熄灭，但是由于飞行中可能突然遭遇乱流，建议在着座时仍确实系好安全带。稍后我们将准备降落，如需要使用洗手间的贵宾，请尽快使用，谢谢。", "chunks": [ { "text": "各位贵宾，虽然现在安全带指示灯已经熄灭，但是由于飞行中可能突然遭遇乱流，建议在着座时仍确实系好安全带。稍后我们将准备降落，如需要使用洗手间的贵宾，请尽快使用，谢谢。", "start": 0.0, "end": 15.8, "chunk_index": 0, "language": "Chinese", "finish_reason": "eos", "truncated": false, "generated_tokens": 53, "max_new_tokens": 222 } ], "detected_language": "Chinese", "segments": [ { "start": 1045.1000000000001, "end": 1060.9, "text": "各位贵宾，虽然现在安全带指示灯已经熄灭，但是由于飞行中可能突然遭遇乱流，建议在着座时仍确实系好安全带。稍后我们将准备降落，如需要使用洗手间的贵宾，请尽快使用，谢谢。" } ], "start": 1045.1000000000001, "end": 1060.8999999999999, "engine": "mlx-qwen3-asr", "language": "zh" }, "raw_word_timestamps": [ { "word": "各位", "start": 1045.7800000000002, "end": 1046.18, "probability": 0.86865234375 }, { "word": "貴", "start": 1046.18, "end": 1046.4, "probability": 0.779296875 }, { "word": "賓,", "start": 1046.4, "end": 1046.6200000000001, "probability": 0.99560546875 }, { "word": "雖", "start": 1046.7600000000002, "end": 1046.9800000000002, "probability": 0.986328125 }, { "word": "然", "start": 1046.9800000000002, "end": 1047.0600000000002, "probability": 1.0 }, { "word": "現在", "start": 1047.0600000000002, "end": 1047.2600000000002, "probability": 0.4091796875 }, { "word": "安全", "start": 1047.2600000000002, "end": 1047.6000000000001, "probability": 0.96826171875 }, { "word": "帶", "start": 1047.6000000000001, "end": 1047.8400000000001, "probability": 0.958984375 }, { "word": "只是", "start": 1047.8400000000001, "end": 1048.1000000000001, "probability": 0.970703125 }, { "word": "在", "start": 1048.1000000000001, "end": 1048.3600000000001, "probability": 0.806640625 }, { "word": "引", "start": 1048.3600000000001, "end": 1048.5600000000002, "probability": 0.50927734375 }, { "word": "進", "start": 1048.5600000000002, "end": 1048.7, "probability": 0.6337890625 }, { "word": "前", "start": 1048.7, "end": 1048.96, "probability": 0.447509765625 }, { "word": "面", "start": 1048.96, "end": 1049.22, "probability": 0.9990234375 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "各位貴賓，雖然現在安全帶指示燈已經熄滅，但是由於飛行中可能突然遭遇亂流，建議在著座時仍確實系好安全帶。", "review_status": "CERTAIN_RESTORATION", "classification": "CERTAIN_RESTORATION", "confidence": 0.95, "correction_reason": "Independent ASR clarifies 'extinguished' and 'turbulence' replacing nonsensical source phrasing." }, { "start": 1049.22, "end": 1051.7, "text": "但是由於飛行中可能突然遭遇亂流", "raw_asr_text": "但是由於飛行中可能突然遭遇浪流", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 68, "words": [], "asr_metrics": { "avg_logprob": -0.28191588906680837, "compression_ratio": 1.1306532663316582, "no_speech_prob": 3.142748214446378e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "各位貴賓,雖然現在安全帶只是在引進前面但是由於飛行中可能突然遭遇浪流簡易在救助時,人確實系好安全帶之後我們將準備降落如需要使用洗手間的貴賓請盡快使用,謝謝" } ], "id": 166, "review_annotation": "", "row_id": 166, "bounded_row_id": 166, "parent_row_ids": [ 166 ], "parent_row_id": 166, "packet_id": 68, "LID_evidence": [ { "region_id": 153, "probabilities": { "ja": 0.005180932303321444, "zh": 0.9626176589065127, "en": 0.023427424383246236 }, "status": "STABLE" } ], "VAD_region_ids": [ 153 ], "language_confidence": 0.9626176589065127, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1045.1000000000001, "packet_end": 1060.8999999999999, "raw_text": "但是由於飛行中可能突然遭遇浪流", "selected_text": "但是由於飛行中可能突然遭遇亂流", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 166 ], "packet_id": 68, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "但是由於飛行中可能突然遭遇浪流", "after": "但是由於飛行中可能突然遭遇亂流", "classification": "CERTAIN_RESTORATION", "reason": "Source '浪流' is a phonetic error for '亂流' (turbulence), confirmed by independent ASR." } ], "language": "zh", "packet_ids": [ 68 ], "index": 166, "independent_ASR": { "text": "各位贵宾，虽然现在安全带指示灯已经熄灭，但是由于飞行中可能突然遭遇乱流，建议在着座时仍确实系好安全带。稍后我们将准备降落，如需要使用洗手间的贵宾，请尽快使用，谢谢。", "chunks": [ { "text": "各位贵宾，虽然现在安全带指示灯已经熄灭，但是由于飞行中可能突然遭遇乱流，建议在着座时仍确实系好安全带。稍后我们将准备降落，如需要使用洗手间的贵宾，请尽快使用，谢谢。", "start": 0.0, "end": 15.8, "chunk_index": 0, "language": "Chinese", "finish_reason": "eos", "truncated": false, "generated_tokens": 53, "max_new_tokens": 222 } ], "detected_language": "Chinese", "segments": [ { "start": 1045.1000000000001, "end": 1060.9, "text": "各位贵宾，虽然现在安全带指示灯已经熄灭，但是由于飞行中可能突然遭遇乱流，建议在着座时仍确实系好安全带。稍后我们将准备降落，如需要使用洗手间的贵宾，请尽快使用，谢谢。" } ], "start": 1045.1000000000001, "end": 1060.8999999999999, "engine": "mlx-qwen3-asr", "language": "zh" }, "raw_word_timestamps": [ { "word": "但是", "start": 1049.22, "end": 1049.5600000000002, "probability": 0.61328125 }, { "word": "由", "start": 1049.5600000000002, "end": 1049.8000000000002, "probability": 0.96923828125 }, { "word": "於", "start": 1049.8000000000002, "end": 1049.92, "probability": 0.99267578125 }, { "word": "飛", "start": 1049.92, "end": 1050.0200000000002, "probability": 0.98583984375 }, { "word": "行", "start": 1050.0200000000002, "end": 1050.18, "probability": 0.9677734375 }, { "word": "中", "start": 1050.18, "end": 1050.38, "probability": 0.9677734375 }, { "word": "可能", "start": 1050.38, "end": 1050.66, "probability": 0.80712890625 }, { "word": "突", "start": 1050.66, "end": 1050.8400000000001, "probability": 0.99462890625 }, { "word": "然", "start": 1050.8400000000001, "end": 1050.96, "probability": 0.99951171875 }, { "word": "遭", "start": 1050.96, "end": 1051.18, "probability": 0.98095703125 }, { "word": "遇", "start": 1051.18, "end": 1051.3000000000002, "probability": 0.98876953125 }, { "word": "浪", "start": 1051.3000000000002, "end": 1051.5000000000002, "probability": 0.8876953125 }, { "word": "流", "start": 1051.5000000000002, "end": 1051.7, "probability": 0.99169921875 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "但是由於飛行中可能突然遭遇亂流", "review_status": "CERTAIN_RESTORATION", "classification": "CERTAIN_RESTORATION", "confidence": 0.95, "correction_reason": "Source '浪流' is a phonetic error for '亂流' (turbulence), confirmed by independent ASR." }, { "start": 1051.7, "end": 1054.3000000000002, "text": "建議在著座時仍確實系好安全帶", "raw_asr_text": "簡易在救助時,人確實系好安全帶", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 68, "words": [], "asr_metrics": { "avg_logprob": -0.28191588906680837, "compression_ratio": 1.1306532663316582, "no_speech_prob": 3.142748214446378e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "各位貴賓,雖然現在安全帶只是在引進前面但是由於飛行中可能突然遭遇浪流簡易在救助時,人確實系好安全帶之後我們將準備降落如需要使用洗手間的貴賓請盡快使用,謝謝" } ], "id": 167, "review_annotation": "", "row_id": 167, "bounded_row_id": 167, "parent_row_ids": [ 167 ], "parent_row_id": 167, "packet_id": 68, "LID_evidence": [ { "region_id": 153, "probabilities": { "ja": 0.005180932303321444, "zh": 0.9626176589065127, "en": 0.023427424383246236 }, "status": "STABLE" } ], "VAD_region_ids": [ 153 ], "language_confidence": 0.9626176589065127, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1045.1000000000001, "packet_end": 1060.8999999999999, "raw_text": "簡易在救助時,人確實系好安全帶", "selected_text": "建議在著座時仍確實系好安全帶", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 167 ], "packet_id": 68, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "簡易在救助時,人確實系好安全帶", "after": "建議在著座時仍確實系好安全帶", "classification": "CERTAIN_RESTORATION", "reason": "Source '簡易在救助時' is nonsense; independent ASR provides the correct safety instruction." } ], "language": "zh", "packet_ids": [ 68 ], "index": 167, "independent_ASR": { "text": "各位贵宾，虽然现在安全带指示灯已经熄灭，但是由于飞行中可能突然遭遇乱流，建议在着座时仍确实系好安全带。稍后我们将准备降落，如需要使用洗手间的贵宾，请尽快使用，谢谢。", "chunks": [ { "text": "各位贵宾，虽然现在安全带指示灯已经熄灭，但是由于飞行中可能突然遭遇乱流，建议在着座时仍确实系好安全带。稍后我们将准备降落，如需要使用洗手间的贵宾，请尽快使用，谢谢。", "start": 0.0, "end": 15.8, "chunk_index": 0, "language": "Chinese", "finish_reason": "eos", "truncated": false, "generated_tokens": 53, "max_new_tokens": 222 } ], "detected_language": "Chinese", "segments": [ { "start": 1045.1000000000001, "end": 1060.9, "text": "各位贵宾，虽然现在安全带指示灯已经熄灭，但是由于飞行中可能突然遭遇乱流，建议在着座时仍确实系好安全带。稍后我们将准备降落，如需要使用洗手间的贵宾，请尽快使用，谢谢。" } ], "start": 1045.1000000000001, "end": 1060.8999999999999, "engine": "mlx-qwen3-asr", "language": "zh" }, "raw_word_timestamps": [ { "word": "簡", "start": 1051.7, "end": 1051.92, "probability": 0.10845947265625 }, { "word": "易", "start": 1051.92, "end": 1052.0600000000002, "probability": 0.833984375 }, { "word": "在", "start": 1052.0600000000002, "end": 1052.2400000000002, "probability": 0.880859375 }, { "word": "救", "start": 1052.2400000000002, "end": 1052.42, "probability": 0.2900390625 }, { "word": "助", "start": 1052.42, "end": 1052.6000000000001, "probability": 0.2076416015625 }, { "word": "時,", "start": 1052.6000000000001, "end": 1052.8400000000001, "probability": 0.9365234375 }, { "word": "人", "start": 1053.0200000000002, "end": 1053.0200000000002, "probability": 0.7529296875 }, { "word": "確", "start": 1053.0200000000002, "end": 1053.22, "probability": 0.9453125 }, { "word": "實", "start": 1053.22, "end": 1053.4, "probability": 0.998046875 }, { "word": "系", "start": 1053.4, "end": 1053.5000000000002, "probability": 0.28466796875 }, { "word": "好", "start": 1053.5000000000002, "end": 1053.66, "probability": 0.9951171875 }, { "word": "安全", "start": 1053.66, "end": 1053.94, "probability": 0.9970703125 }, { "word": "帶", "start": 1053.94, "end": 1054.3000000000002, "probability": 0.99560546875 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "建議在著座時仍確實系好安全帶", "review_status": "CERTAIN_RESTORATION", "classification": "CERTAIN_RESTORATION", "confidence": 0.95, "correction_reason": "Source '簡易在救助時' is nonsense; independent ASR provides the correct safety instruction." }, { "start": 1055.3400000000001, "end": 1056.9800000000002, "text": "之後我們將準備降落", "raw_asr_text": "之後我們將準備降落", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 68, "words": [ { "word": "之後", "start": 1055.3400000000001, "end": 1055.7400000000002, "probability": 0.378173828125 }, { "word": "我們", "start": 1055.7400000000002, "end": 1055.9800000000002, "probability": 0.91064453125 }, { "word": "將", "start": 1055.9800000000002, "end": 1056.16, "probability": 0.978515625 }, { "word": "準備", "start": 1056.16, "end": 1056.5200000000002, "probability": 0.59765625 }, { "word": "降", "start": 1056.5200000000002, "end": 1056.7600000000002, "probability": 0.986328125 }, { "word": "落", "start": 1056.7600000000002, "end": 1056.9800000000002, "probability": 0.99951171875 } ], "asr_metrics": { "avg_logprob": -0.28191588906680837, "compression_ratio": 1.1306532663316582, "no_speech_prob": 3.142748214446378e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "各位貴賓,雖然現在安全帶只是在引進前面但是由於飛行中可能突然遭遇浪流簡易在救助時,人確實系好安全帶之後我們將準備降落如需要使用洗手間的貴賓請盡快使用,謝謝" } ], "id": 168, "review_annotation": "", "row_id": 168, "bounded_row_id": 168, "parent_row_ids": [ 168 ], "parent_row_id": 168, "packet_id": 68, "LID_evidence": [ { "region_id": 154, "probabilities": { "ja": 0.003708197321328852, "zh": 0.9675877425405714, "en": 0.02204172491716842 }, "status": "STABLE" } ], "VAD_region_ids": [ 154 ], "language_confidence": 0.9675877425405714, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1045.1000000000001, "packet_end": 1060.8999999999999, "raw_text": "之後我們將準備降落", "selected_text": "之後我們將準備降落", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 168 ], "packet_id": 68, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "之後我們將準備降落", "after": "之後我們將準備降落", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard phrase matching the Japanese '着陸体制に入ります' context." } ], "language": "zh", "packet_ids": [ 68 ], "index": 168, "source_text": "之後我們將準備降落", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Standard phrase matching the Japanese '着陸体制に入ります' context." }, { "start": 1056.9800000000002, "end": 1059.64, "text": "如需要使用洗手間的貴賓請盡快使用,謝謝", "raw_asr_text": "如需要使用洗手間的貴賓請盡快使用,謝謝", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 68, "words": [ { "word": "如", "start": 1056.9800000000002, "end": 1057.2, "probability": 0.6337890625 }, { "word": "需要", "start": 1057.2, "end": 1057.4, "probability": 0.9306640625 }, { "word": "使", "start": 1057.4, "end": 1057.6000000000001, "probability": 0.94384765625 }, { "word": "用", "start": 1057.6000000000001, "end": 1057.72, "probability": 0.99951171875 }, { "word": "洗", "start": 1057.72, "end": 1057.8200000000002, "probability": 0.99658203125 }, { "word": "手", "start": 1057.8200000000002, "end": 1057.96, "probability": 0.99951171875 }, { "word": "間", "start": 1057.96, "end": 1058.0800000000002, "probability": 0.93017578125 }, { "word": "的", "start": 1058.0800000000002, "end": 1058.22, "probability": 0.99267578125 }, { "word": "貴", "start": 1058.22, "end": 1058.38, "probability": 0.9990234375 }, { "word": "賓", "start": 1058.38, "end": 1058.5200000000002, "probability": 1.0 }, { "word": "請", "start": 1058.5200000000002, "end": 1058.7, "probability": 0.50244140625 }, { "word": "盡", "start": 1058.7, "end": 1058.8600000000001, "probability": 0.8203125 }, { "word": "快", "start": 1058.8600000000001, "end": 1059.0200000000002, "probability": 0.99755859375 }, { "word": "使", "start": 1059.0200000000002, "end": 1059.22, "probability": 0.984375 }, { "word": "用,", "start": 1059.22, "end": 1059.4, "probability": 1.0 }, { "word": "謝謝", "start": 1059.5000000000002, "end": 1059.64, "probability": 0.990234375 } ], "asr_metrics": { "avg_logprob": -0.28191588906680837, "compression_ratio": 1.1306532663316582, "no_speech_prob": 3.142748214446378e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "各位貴賓,雖然現在安全帶只是在引進前面但是由於飛行中可能突然遭遇浪流簡易在救助時,人確實系好安全帶之後我們將準備降落如需要使用洗手間的貴賓請盡快使用,謝謝" } ], "id": 169, "review_annotation": "", "row_id": 169, "bounded_row_id": 169, "parent_row_ids": [ 169 ], "parent_row_id": 169, "packet_id": 68, "LID_evidence": [ { "region_id": 154, "probabilities": { "ja": 0.003708197321328852, "zh": 0.9675877425405714, "en": 0.02204172491716842 }, "status": "STABLE" } ], "VAD_region_ids": [ 154 ], "language_confidence": 0.9675877425405714, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1045.1000000000001, "packet_end": 1060.8999999999999, "raw_text": "如需要使用洗手間的貴賓請盡快使用,謝謝", "selected_text": "如需要使用洗手間的貴賓請盡快使用,謝謝", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 169 ], "packet_id": 68, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "如需要使用洗手間的貴賓請盡快使用,謝謝", "after": "如需要使用洗手間的貴賓請盡快使用,謝謝", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard instruction matching the Japanese lavatory announcement." } ], "language": "zh", "packet_ids": [ 68 ], "index": 169, "source_text": "如需要使用洗手間的貴賓請盡快使用,謝謝", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard instruction matching the Japanese lavatory announcement." }, { "start": 1062.42, "end": 1067.0400000000002, "text": "皆様、シートベルトは緩みのないようしっかりお締めください。", "raw_asr_text": "皆様、シートベルトは緩みのないようしっかりお締めください。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 69, "words": [ { "word": "皆", "start": 1062.42, "end": 1062.7400000000002, "probability": 0.89599609375 }, { "word": "様、", "start": 1062.7400000000002, "end": 1063.0000000000002, "probability": 0.611328125 }, { "word": "シ", "start": 1063.5000000000002, "end": 1063.96, "probability": 0.99951171875 }, { "word": "ート", "start": 1063.96, "end": 1064.14, "probability": 1.0 }, { "word": "ベ", "start": 1064.14, "end": 1064.3000000000002, "probability": 0.99951171875 }, { "word": "ル", "start": 1064.3000000000002, "end": 1064.38, "probability": 0.99951171875 }, { "word": "ト", "start": 1064.38, "end": 1064.5000000000002, "probability": 0.99951171875 }, { "word": "は", "start": 1064.5000000000002, "end": 1064.68, "probability": 1.0 }, { "word": "緩", "start": 1064.68, "end": 1065.22, "probability": 0.9755859375 }, { "word": "み", "start": 1065.22, "end": 1065.3600000000001, "probability": 1.0 }, { "word": "の", "start": 1065.3600000000001, "end": 1065.4800000000002, "probability": 1.0 }, { "word": "ない", "start": 1065.4800000000002, "end": 1065.68, "probability": 0.99951171875 }, { "word": "よう", "start": 1065.68, "end": 1065.8400000000001, "probability": 0.9912109375 }, { "word": "し", "start": 1065.8400000000001, "end": 1066.0400000000002, "probability": 0.8486328125 }, { "word": "っ", "start": 1066.0400000000002, "end": 1066.14, "probability": 1.0 }, { "word": "か", "start": 1066.14, "end": 1066.2400000000002, "probability": 1.0 }, { "word": "り", "start": 1066.2400000000002, "end": 1066.3400000000001, "probability": 1.0 }, { "word": "お", "start": 1066.3400000000001, "end": 1066.46, "probability": 0.986328125 }, { "word": "締", "start": 1066.46, "end": 1066.6200000000001, "probability": 0.9931640625 }, { "word": "め", "start": 1066.6200000000001, "end": 1066.7, "probability": 1.0 }, { "word": "ください。", "start": 1066.7, "end": 1067.0400000000002, "probability": 0.994140625 } ], "asr_metrics": { "avg_logprob": -0.13882357733590261, "compression_ratio": 1.0481927710843373, "no_speech_prob": 4.32468297284494e-12 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "皆様、シートベルトは緩みのないようしっかりお締めください。" } ], "id": 170, "review_annotation": "", "row_id": 170, "bounded_row_id": 170, "parent_row_ids": [ 170 ], "parent_row_id": 170, "packet_id": 69, "LID_evidence": [ { "region_id": 155, "probabilities": { "ja": 0.991256445646286, "zh": 0.0004323240937083028, "en": 0.004698505596024916 }, "status": "UNCERTAIN" }, { "region_id": 156, "probabilities": { "ja": 0.9777384102344513, "zh": 0.001101240669231629, "en": 0.011255347909172996 }, "status": "STABLE" } ], "VAD_region_ids": [ 155, 156 ], "language_confidence": 0.991256445646286, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1061.6000000000001, "packet_end": 1068.2, "raw_text": "皆様、シートベルトは緩みのないようしっかりお締めください。", "selected_text": "皆様、シートベルトは緩みのないようしっかりお締めください。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 170 ], "packet_id": 69, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "皆様、シートベルトは緩みのないようしっかりお締めください。", "after": "皆様、シートベルトは緩みのないようしっかりお締めください。", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard safety announcement phrasing matches context and parallel text perfectly." } ], "language": "ja", "packet_ids": [ 69 ], "index": 170, "source_text": "皆様、シートベルトは緩みのないようしっかりお締めください。", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard safety announcement phrasing matches context and parallel text perfectly." }, { "start": 1068.3, "end": 1073.04, "text": "Ladies and gentlemen, please make sure that your seatbelt is securely fastened.", "raw_asr_text": "Ladies and gentlemen, please make sure that your seatbelt is securely fastened.", "source_language": "en", "source_lang": "en", "asr_packet_id": 70, "words": [], "asr_metrics": { "avg_logprob": -0.33379562695821124, "compression_ratio": 1.0843373493975903, "no_speech_prob": 1.5661332907618286e-12 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "en", "text": "Ladies and gentlemen, please make sure that your seatbelt is securely fastened. Thank you." }, { "language": "ja", "text": "ご視聴ありがとうございました" } ], "id": 171, "review_annotation": "", "row_id": 171, "bounded_row_id": 171, "parent_row_ids": [ 171 ], "parent_row_id": 171, "packet_id": 70, "LID_evidence": [ { "region_id": 157, "probabilities": { "ja": 0.2631964128855283, "zh": 0.001523889377235069, "en": 0.7198777280614844 }, "status": "STABLE" }, { "region_id": 158, "probabilities": { "ja": 0.000173183837129424, "zh": 0.00010789294244912323, "en": 0.9985332157876756 }, "status": "STABLE" } ], "VAD_region_ids": [ 157, 158 ], "language_confidence": 0.9985332157876756, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1067.5, "packet_end": 1075.1, "raw_text": "Ladies and gentlemen, please make sure that your seatbelt is securely fastened.", "selected_text": "Ladies and gentlemen, please make sure that your seatbelt is securely fastened.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 171 ], "packet_id": 70, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "Ladies and gentlemen, please make sure that your seatbelt is securely fastened.", "after": "Ladies and gentlemen, please make sure that your seatbelt is securely fastened.", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard English safety announcement matches context and parallel text perfectly." } ], "language": "en", "packet_ids": [ 70 ], "index": 171, "source_text": "Ladies and gentlemen, please make sure that your seatbelt is securely fastened.", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard English safety announcement matches context and parallel text perfectly.", "raw_word_timestamps": [ { "word": " Ladies", "start": 1067.5, "end": 1068.72, "probability": 0.56982421875 }, { "word": " and", "start": 1068.72, "end": 1068.98, "probability": 0.99560546875 }, { "word": " gentlemen,", "start": 1068.98, "end": 1069.36, "probability": 0.86376953125 }, { "word": " please", "start": 1069.82, "end": 1070.38, "probability": 0.9765625 }, { "word": " make", "start": 1070.38, "end": 1070.6, "probability": 0.99755859375 }, { "word": " sure", "start": 1070.6, "end": 1070.92, "probability": 0.998046875 }, { "word": " that", "start": 1070.92, "end": 1071.16, "probability": 0.98388671875 }, { "word": " your", "start": 1071.16, "end": 1071.3, "probability": 0.99462890625 }, { "word": " seatbelt", "start": 1071.3, "end": 1071.74, "probability": 0.8134765625 }, { "word": " is", "start": 1071.74, "end": 1071.9, "probability": 0.99951171875 }, { "word": " securely", "start": 1071.9, "end": 1072.32, "probability": 0.99658203125 }, { "word": " fastened.", "start": 1072.32, "end": 1073.04, "probability": 0.9990234375 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 1073.24, "end": 1073.96, "text": "Thank you.", "raw_asr_text": "Thank you.", "source_language": "en", "source_lang": "en", "asr_packet_id": 70, "words": [ { "word": " Thank", "start": 1073.24, "end": 1073.68, "probability": 0.9990234375 }, { "word": " you.", "start": 1073.68, "end": 1073.96, "probability": 0.998046875 } ], "asr_metrics": { "avg_logprob": -0.33379562695821124, "compression_ratio": 1.0843373493975903, "no_speech_prob": 1.5661332907618286e-12 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "en", "text": "Ladies and gentlemen, please make sure that your seatbelt is securely fastened. Thank you." }, { "language": "ja", "text": "ご視聴ありがとうございました" } ], "id": 172, "review_annotation": "", "row_id": 172, "bounded_row_id": 172, "parent_row_ids": [ 172 ], "parent_row_id": 172, "packet_id": 70, "LID_evidence": [ { "region_id": 158, "probabilities": { "ja": 0.000173183837129424, "zh": 0.00010789294244912323, "en": 0.9985332157876756 }, "status": "STABLE" }, { "region_id": 159, "probabilities": { "ja": 8.526765850547235e-05, "zh": 5.481831567522022e-05, "en": 0.999019205570221 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 158, 159 ], "language_confidence": 0.999019205570221, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1067.5, "packet_end": 1075.1, "raw_text": "Thank you.", "selected_text": "Thank you.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 172 ], "packet_id": 70, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "Thank you.", "after": "Thank you.", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard closing phrase, consistent with surrounding context." } ], "language": "en", "packet_ids": [ 70 ], "index": 172, "source_text": "Thank you.", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard closing phrase, consistent with surrounding context." }, { "start": 1076.8, "end": 1079.14, "text": "皆様、着陸体制に入りました。", "raw_asr_text": "皆様、着陸体制に入りました。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 71, "words": [], "asr_metrics": { "avg_logprob": -0.12474610084711119, "compression_ratio": 1.4293193717277486, "no_speech_prob": 7.895621656484053e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "皆様、着陸体制に入りました。シートベルトを緩みのないようしっかりお締めください。座席のリクライニングとテーブル、フットレスト、シートテレビ、コントローラーを元の位置にお戻しください。" } ], "id": 173, "review_annotation": "", "row_id": 173, "bounded_row_id": 173, "parent_row_ids": [ 173 ], "parent_row_id": 173, "packet_id": 71, "LID_evidence": [ { "region_id": 160, "probabilities": { "ja": 0.9937799374262491, "zh": 0.0003693783995307361, "en": 0.0030260743224062026 }, "status": "STABLE" } ], "VAD_region_ids": [ 160 ], "language_confidence": 0.9937799374262491, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1076.0, "packet_end": 1093.2, "raw_text": "皆様、着陸体制に入りました。", "selected_text": "皆様、着陸体制に入りました。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 173 ], "packet_id": 71, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "皆様、着陸体制に入りました。", "after": "皆様、着陸体制に入りました。", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard announcement phrase, consistent with context." } ], "language": "ja", "packet_ids": [ 71 ], "index": 173, "source_text": "皆様、着陸体制に入りました。", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard announcement phrase, consistent with context.", "raw_word_timestamps": [ { "word": "皆", "start": 1076.76, "end": 1077.08, "probability": 0.845703125 }, { "word": "様、", "start": 1077.08, "end": 1077.34, "probability": 0.52734375 }, { "word": "着", "start": 1077.5, "end": 1077.64, "probability": 1.0 }, { "word": "陸", "start": 1077.64, "end": 1077.9, "probability": 0.99951171875 }, { "word": "体", "start": 1077.9, "end": 1078.08, "probability": 0.61083984375 }, { "word": "制", "start": 1078.08, "end": 1078.28, "probability": 0.96484375 }, { "word": "に", "start": 1078.28, "end": 1078.46, "probability": 1.0 }, { "word": "入", "start": 1078.46, "end": 1078.68, "probability": 1.0 }, { "word": "りました。", "start": 1078.68, "end": 1079.14, "probability": 1.0 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 1079.9, "end": 1083.02, "text": "シートベルトを緩みのないようしっかりお締めください。", "raw_asr_text": "シートベルトを緩みのないようしっかりお締めください。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 71, "words": [ { "word": "シ", "start": 1079.9, "end": 1080.22, "probability": 0.98388671875 }, { "word": "ート", "start": 1080.22, "end": 1080.36, "probability": 1.0 }, { "word": "ベ", "start": 1080.36, "end": 1080.54, "probability": 0.99951171875 }, { "word": "ル", "start": 1080.54, "end": 1080.62, "probability": 0.99951171875 }, { "word": "ト", "start": 1080.62, "end": 1080.76, "probability": 0.99951171875 }, { "word": "を", "start": 1080.76, "end": 1081.0, "probability": 0.9853515625 }, { "word": "緩", "start": 1081.0, "end": 1081.22, "probability": 0.95654296875 }, { "word": "み", "start": 1081.22, "end": 1081.34, "probability": 0.9990234375 }, { "word": "の", "start": 1081.34, "end": 1081.48, "probability": 0.99951171875 }, { "word": "ない", "start": 1081.48, "end": 1081.68, "probability": 0.98876953125 }, { "word": "よう", "start": 1081.68, "end": 1081.84, "probability": 0.85205078125 }, { "word": "し", "start": 1081.84, "end": 1082.06, "probability": 0.6455078125 }, { "word": "っ", "start": 1082.06, "end": 1082.12, "probability": 1.0 }, { "word": "か", "start": 1082.12, "end": 1082.18, "probability": 1.0 }, { "word": "り", "start": 1082.18, "end": 1082.28, "probability": 1.0 }, { "word": "お", "start": 1082.28, "end": 1082.38, "probability": 0.9912109375 }, { "word": "締", "start": 1082.38, "end": 1082.5, "probability": 0.97265625 }, { "word": "め", "start": 1082.5, "end": 1082.6, "probability": 0.99951171875 }, { "word": "ください。", "start": 1082.6, "end": 1083.02, "probability": 0.9912109375 } ], "asr_metrics": { "avg_logprob": -0.12474610084711119, "compression_ratio": 1.4293193717277486, "no_speech_prob": 7.895621656484053e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "皆様、着陸体制に入りました。シートベルトを緩みのないようしっかりお締めください。座席のリクライニングとテーブル、フットレスト、シートテレビ、コントローラーを元の位置にお戻しください。" } ], "id": 174, "review_annotation": "", "row_id": 174, "bounded_row_id": 174, "parent_row_ids": [ 174 ], "parent_row_id": 174, "packet_id": 71, "LID_evidence": [ { "region_id": 161, "probabilities": { "ja": 0.990480974316597, "zh": 0.000514746021508472, "en": 0.004277100757462904 }, "status": "STABLE" } ], "VAD_region_ids": [ 161 ], "language_confidence": 0.990480974316597, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1076.0, "packet_end": 1093.2, "raw_text": "シートベルトを緩みのないようしっかりお締めください。", "selected_text": "シートベルトを緩みのないようしっかりお締めください。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 174 ], "packet_id": 71, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "シートベルトを緩みのないようしっかりお締めください。", "after": "シートベルトを緩みのないようしっかりお締めください。", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard safety instruction, consistent with context." } ], "language": "ja", "packet_ids": [ 71 ], "index": 174, "source_text": "シートベルトを緩みのないようしっかりお締めください。", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard safety instruction, consistent with context." }, { "start": 1084.82, "end": 1091.82, "text": "座席のリクライニングとテーブル、フットレスト、シートテレビ、コントローラーを元の位置にお戻しください。", "raw_asr_text": "座席のリクライニングとテーブル、フットレスト、シートテレビ、コントローラーを元の位置にお戻しください。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 71, "words": [ { "word": "座", "start": 1084.82, "end": 1085.08, "probability": 0.9580078125 }, { "word": "席", "start": 1085.08, "end": 1085.24, "probability": 0.947265625 }, { "word": "の", "start": 1085.24, "end": 1085.4, "probability": 0.9990234375 }, { "word": "リ", "start": 1085.4, "end": 1085.52, "probability": 0.99853515625 }, { "word": "ク", "start": 1085.52, "end": 1085.62, "probability": 0.99951171875 }, { "word": "ライ", "start": 1085.62, "end": 1085.8, "probability": 1.0 }, { "word": "ニ", "start": 1085.8, "end": 1085.96, "probability": 0.99609375 }, { "word": "ング", "start": 1085.96, "end": 1086.1, "probability": 1.0 }, { "word": "と", "start": 1086.1, "end": 1086.28, "probability": 0.98681640625 }, { "word": "テ", "start": 1086.28, "end": 1086.5, "probability": 0.99853515625 }, { "word": "ーブ", "start": 1086.5, "end": 1086.64, "probability": 1.0 }, { "word": "ル、", "start": 1086.64, "end": 1086.78, "probability": 1.0 }, { "word": "フ", "start": 1087.08, "end": 1087.4, "probability": 0.9951171875 }, { "word": "ット", "start": 1087.4, "end": 1087.56, "probability": 0.998046875 }, { "word": "レ", "start": 1087.56, "end": 1087.74, "probability": 0.99609375 }, { "word": "スト、", "start": 1087.74, "end": 1087.92, "probability": 0.85546875 }, { "word": "シ", "start": 1088.46, "end": 1088.64, "probability": 0.99609375 }, { "word": "ート", "start": 1088.64, "end": 1088.82, "probability": 1.0 }, { "word": "テ", "start": 1088.82, "end": 1088.98, "probability": 0.99951171875 }, { "word": "レ", "start": 1088.98, "end": 1089.08, "probability": 1.0 }, { "word": "ビ、", "start": 1089.08, "end": 1089.22, "probability": 0.99951171875 }, { "word": "コ", "start": 1089.48, "end": 1089.6, "probability": 0.99609375 }, { "word": "ント", "start": 1089.6, "end": 1089.7, "probability": 1.0 }, { "word": "ロ", "start": 1089.7, "end": 1089.86, "probability": 1.0 }, { "word": "ー", "start": 1089.86, "end": 1089.96, "probability": 0.99951171875 }, { "word": "ラ", "start": 1089.96, "end": 1090.06, "probability": 1.0 }, { "word": "ー", "start": 1090.06, "end": 1090.14, "probability": 0.99267578125 }, { "word": "を", "start": 1090.14, "end": 1090.28, "probability": 0.99951171875 }, { "word": "元", "start": 1090.28, "end": 1090.54, "probability": 0.99462890625 }, { "word": "の", "start": 1090.54, "end": 1090.68, "probability": 1.0 }, { "word": "位", "start": 1090.68, "end": 1090.86, "probability": 0.99951171875 }, { "word": "置", "start": 1090.86, "end": 1090.92, "probability": 1.0 }, { "word": "に", "start": 1090.92, "end": 1091.06, "probability": 1.0 }, { "word": "お", "start": 1091.06, "end": 1091.22, "probability": 0.990234375 }, { "word": "戻", "start": 1091.22, "end": 1091.38, "probability": 0.9990234375 }, { "word": "し", "start": 1091.38, "end": 1091.48, "probability": 0.9990234375 }, { "word": "ください。", "start": 1091.48, "end": 1091.82, "probability": 0.99951171875 } ], "asr_metrics": { "avg_logprob": -0.12474610084711119, "compression_ratio": 1.4293193717277486, "no_speech_prob": 7.895621656484053e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "皆様、着陸体制に入りました。シートベルトを緩みのないようしっかりお締めください。座席のリクライニングとテーブル、フットレスト、シートテレビ、コントローラーを元の位置にお戻しください。" } ], "id": 175, "review_annotation": "", "row_id": 175, "bounded_row_id": 175, "parent_row_ids": [ 175 ], "parent_row_id": 175, "packet_id": 71, "LID_evidence": [ { "region_id": 162, "probabilities": { "ja": 0.9724297065001267, "zh": 0.0014026693847531883, "en": 0.014033543841483502 }, "status": "STABLE" } ], "VAD_region_ids": [ 162 ], "language_confidence": 0.9724297065001267, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1076.0, "packet_end": 1093.2, "raw_text": "座席のリクライニングとテーブル、フットレスト、シートテレビ、コントローラーを元の位置にお戻しください。", "selected_text": "座席のリクライニングとテーブル、フットレスト、シートテレビ、コントローラーを元の位置にお戻しください。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 175 ], "packet_id": 71, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "座席のリクライニングとテーブル、フットレスト、シートテレビ、コントローラーを元の位置にお戻しください。", "after": "座席のリクライニングとテーブル、フットレスト、シートテレビ、コントローラーを元の位置にお戻しください。", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard pre-landing instruction, matches context and parallel text." } ], "language": "ja", "packet_ids": [ 71 ], "index": 175, "source_text": "座席のリクライニングとテーブル、フットレスト、シートテレビ、コントローラーを元の位置にお戻しください。", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard pre-landing instruction, matches context and parallel text." }, { "start": 1095.9, "end": 1100.3600000000001, "text": "手荷物は上の物入れ、エコノミークラスは前の座席の下にお入れください。", "raw_asr_text": "手荷物は上の物入れ、エコノミークラスは前の座席の下にお入れください。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 72, "words": [], "asr_metrics": { "avg_logprob": -0.1048179122636903, "compression_ratio": 1.2096774193548387, "no_speech_prob": 6.125152902242137e-12 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "手荷物は上の物入れ、エコノミークラスは前の座席の下にお入れください。化粧室のご使用はお控えください。" } ], "id": 176, "review_annotation": "", "row_id": 176, "bounded_row_id": 176, "parent_row_ids": [ 176 ], "parent_row_id": 176, "packet_id": 72, "LID_evidence": [ { "region_id": 163, "probabilities": { "ja": 0.973429126398904, "zh": 0.0018365898660184549, "en": 0.012678639547500228 }, "status": "STABLE" } ], "VAD_region_ids": [ 163 ], "language_confidence": 0.973429126398904, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1095.1000000000001, "packet_end": 1105.8, "raw_text": "手荷物は上の物入れ、エコノミークラスは前の座席の下にお入れください。", "selected_text": "手荷物は上の物入れ、エコノミークラスは前の座席の下にお入れください。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 176 ], "packet_id": 72, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "手荷物は上の物入れ、エコノミークラスは前の座席の下にお入れください。", "after": "手荷物は上の物入れ、エコノミークラスは前の座席の下にお入れください。", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard baggage instruction, matches context and parallel text." } ], "language": "ja", "packet_ids": [ 72 ], "index": 176, "source_text": "手荷物は上の物入れ、エコノミークラスは前の座席の下にお入れください。", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard baggage instruction, matches context and parallel text.", "raw_word_timestamps": [ { "word": "手", "start": 1095.8000000000002, "end": 1096.1200000000001, "probability": 0.91357421875 }, { "word": "荷", "start": 1096.1200000000001, "end": 1096.2400000000002, "probability": 0.9970703125 }, { "word": "物", "start": 1096.2400000000002, "end": 1096.4, "probability": 0.99951171875 }, { "word": "は", "start": 1096.4, "end": 1096.6000000000001, "probability": 0.99951171875 }, { "word": "上", "start": 1096.6000000000001, "end": 1096.8000000000002, "probability": 0.984375 }, { "word": "の", "start": 1096.8000000000002, "end": 1096.96, "probability": 0.9990234375 }, { "word": "物", "start": 1096.96, "end": 1097.16, "probability": 0.9697265625 }, { "word": "入", "start": 1097.16, "end": 1097.3600000000001, "probability": 0.9658203125 }, { "word": "れ、", "start": 1097.3600000000001, "end": 1097.4800000000002, "probability": 0.99951171875 }, { "word": "エ", "start": 1097.5800000000002, "end": 1097.7, "probability": 0.97705078125 }, { "word": "コ", "start": 1097.7, "end": 1097.7800000000002, "probability": 0.9990234375 }, { "word": "ノ", "start": 1097.7800000000002, "end": 1097.8600000000001, "probability": 0.990234375 }, { "word": "ミー", "start": 1097.8600000000001, "end": 1098.0000000000002, "probability": 0.97509765625 }, { "word": "ク", "start": 1098.0000000000002, "end": 1098.16, "probability": 0.99609375 }, { "word": "ラ", "start": 1098.16, "end": 1098.2600000000002, "probability": 0.99951171875 }, { "word": "ス", "start": 1098.2600000000002, "end": 1098.3600000000001, "probability": 0.99951171875 }, { "word": "は", "start": 1098.3600000000001, "end": 1098.5000000000002, "probability": 1.0 }, { "word": "前", "start": 1098.5000000000002, "end": 1098.66, "probability": 0.99853515625 }, { "word": "の", "start": 1098.66, "end": 1098.7800000000002, "probability": 1.0 }, { "word": "座", "start": 1098.7800000000002, "end": 1098.9, "probability": 0.9990234375 }, { "word": "席", "start": 1098.9, "end": 1099.1000000000001, "probability": 1.0 }, { "word": "の", "start": 1099.1000000000001, "end": 1099.2400000000002, "probability": 1.0 }, { "word": "下", "start": 1099.2400000000002, "end": 1099.44, "probability": 1.0 }, { "word": "に", "start": 1099.44, "end": 1099.64, "probability": 1.0 }, { "word": "お", "start": 1099.64, "end": 1099.8000000000002, "probability": 0.900390625 }, { "word": "入", "start": 1099.8000000000002, "end": 1099.8600000000001, "probability": 0.998046875 }, { "word": "れ", "start": 1099.8600000000001, "end": 1099.92, "probability": 0.99951171875 }, { "word": "ください。", "start": 1099.92, "end": 1100.3600000000001, "probability": 0.998046875 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 1102.3200000000002, "end": 1104.46, "text": "化粧室のご使用はお控えください。", "raw_asr_text": "化粧室のご使用はお控えください。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 72, "words": [ { "word": "化", "start": 1102.3200000000002, "end": 1102.64, "probability": 0.98681640625 }, { "word": "粧", "start": 1102.64, "end": 1102.8200000000002, "probability": 0.9990234375 }, { "word": "室", "start": 1102.8200000000002, "end": 1102.9800000000002, "probability": 0.9970703125 }, { "word": "の", "start": 1102.9800000000002, "end": 1103.1000000000001, "probability": 1.0 }, { "word": "ご", "start": 1103.1000000000001, "end": 1103.2400000000002, "probability": 0.9921875 }, { "word": "使", "start": 1103.2400000000002, "end": 1103.38, "probability": 0.9990234375 }, { "word": "用", "start": 1103.38, "end": 1103.5000000000002, "probability": 1.0 }, { "word": "は", "start": 1103.5000000000002, "end": 1103.66, "probability": 1.0 }, { "word": "お", "start": 1103.66, "end": 1103.8000000000002, "probability": 0.99609375 }, { "word": "控", "start": 1103.8000000000002, "end": 1103.96, "probability": 1.0 }, { "word": "え", "start": 1103.96, "end": 1104.0800000000002, "probability": 1.0 }, { "word": "ください。", "start": 1104.0800000000002, "end": 1104.46, "probability": 0.99951171875 } ], "asr_metrics": { "avg_logprob": -0.1048179122636903, "compression_ratio": 1.2096774193548387, "no_speech_prob": 6.125152902242137e-12 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "手荷物は上の物入れ、エコノミークラスは前の座席の下にお入れください。化粧室のご使用はお控えください。" } ], "id": 177, "review_annotation": "", "row_id": 177, "bounded_row_id": 177, "parent_row_ids": [ 177 ], "parent_row_id": 177, "packet_id": 72, "LID_evidence": [ { "region_id": 164, "probabilities": { "ja": 0.9967979192733765, "zh": 0.00020096139269298874, "en": 0.0017513288767077029 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 164 ], "language_confidence": 0.9967979192733765, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1095.1000000000001, "packet_end": 1105.8, "raw_text": "化粧室のご使用はお控えください。", "selected_text": "化粧室のご使用はお控えください。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 177 ], "packet_id": 72, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "化粧室のご使用はお控えください。", "after": "化粧室のご使用はお控えください。", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard lavatory restriction, matches context and parallel text." } ], "language": "ja", "packet_ids": [ 72 ], "index": 177, "source_text": "化粧室のご使用はお控えください。", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard lavatory restriction, matches context and parallel text." }, { "start": 1108.1, "end": 1111.3, "text": "Ladies and gentlemen, we are now making our final approach.", "raw_asr_text": "Ladies and gentlemen, we are now making our final approach.", "source_language": "en", "source_lang": "en", "asr_packet_id": 73, "words": [], "asr_metrics": { "avg_logprob": -0.27808494567871095, "compression_ratio": 1.4256756756756757, "no_speech_prob": 6.46698389283884e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "en", "text": "Ladies and gentlemen, we are now making our final approach. Please make sure that your seat belt is securely fastened and will turn your seat on table, footless, seat TV and controllers to the original position." } ], "id": 178, "review_annotation": "", "row_id": 178, "bounded_row_id": 178, "parent_row_ids": [ 178 ], "parent_row_id": 178, "packet_id": 73, "LID_evidence": [ { "region_id": 165, "probabilities": { "ja": 0.003709702730702702, "zh": 0.0009958170267054811, "en": 0.9855961576104164 }, "status": "STABLE" } ], "VAD_region_ids": [ 165 ], "language_confidence": 0.9855961576104164, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1107.3, "packet_end": 1121.8999999999999, "raw_text": "Ladies and gentlemen, we are now making our final approach.", "selected_text": "Ladies and gentlemen, we are now making our final approach.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 178 ], "packet_id": 73, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "Ladies and gentlemen, we are now making our final approach.", "after": "Ladies and gentlemen, we are now making our final approach.", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard announcement phrase, matches context and parallel text." } ], "language": "en", "packet_ids": [ 73 ], "index": 178, "source_text": "Ladies and gentlemen, we are now making our final approach.", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard announcement phrase, matches context and parallel text.", "raw_word_timestamps": [ { "word": " Ladies", "start": 1107.98, "end": 1108.54, "probability": 0.8291015625 }, { "word": " and", "start": 1108.54, "end": 1108.82, "probability": 0.9912109375 }, { "word": " gentlemen,", "start": 1108.82, "end": 1109.1399999999999, "probability": 0.81298828125 }, { "word": " we", "start": 1109.32, "end": 1109.3999999999999, "probability": 0.97509765625 }, { "word": " are", "start": 1109.3999999999999, "end": 1109.52, "probability": 0.98486328125 }, { "word": " now", "start": 1109.52, "end": 1109.68, "probability": 0.94287109375 }, { "word": " making", "start": 1109.68, "end": 1110.0, "probability": 0.998046875 }, { "word": " our", "start": 1110.0, "end": 1110.34, "probability": 0.97607421875 }, { "word": " final", "start": 1110.34, "end": 1110.78, "probability": 0.998046875 }, { "word": " approach.", "start": 1110.78, "end": 1111.3, "probability": 0.9169921875 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 1111.6399999999999, "end": 1116.6599999999999, "text": "Please make sure that your seat belt is securely fastened and return your seat, table,", "raw_asr_text": "Please make sure that your seat belt is securely fastened and will turn your seat on table,", "source_language": "en", "source_lang": "en", "asr_packet_id": 73, "words": [], "asr_metrics": { "avg_logprob": -0.27808494567871095, "compression_ratio": 1.4256756756756757, "no_speech_prob": 6.46698389283884e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "en", "text": "Ladies and gentlemen, we are now making our final approach. Please make sure that your seat belt is securely fastened and will turn your seat on table, footless, seat TV and controllers to the original position." } ], "id": 179, "review_annotation": "", "row_id": 179, "bounded_row_id": 179, "parent_row_ids": [ 179 ], "parent_row_id": 179, "packet_id": 73, "LID_evidence": [ { "region_id": 165, "probabilities": { "ja": 0.003709702730702702, "zh": 0.0009958170267054811, "en": 0.9855961576104164 }, "status": "STABLE" }, { "region_id": 166, "probabilities": { "ja": 0.00820494735671673, "zh": 0.01211459712976648, "en": 0.9185813516378403 }, "status": "STABLE" } ], "VAD_region_ids": [ 165, 166 ], "language_confidence": 0.9855961576104164, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1107.3, "packet_end": 1121.8999999999999, "raw_text": "Please make sure that your seat belt is securely fastened and will turn your seat on table,", "selected_text": "Please make sure that your seat belt is securely fastened and return your seat, table,", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 179 ], "packet_id": 73, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "Please make sure that your seat belt is securely fastened and will turn your seat on table,", "after": "Please make sure that your seat belt is securely fastened and return your seat, table,", "classification": "CERTAIN_RESTORATION", "reason": "Source 'will turn' is incorrect; independent ASR confirms 'return' and lists items correctly." } ], "language": "en", "packet_ids": [ 73 ], "index": 179, "independent_ASR": { "text": "Ladies and gentlemen, we are now making our final approach. Please make sure that your seatbelt is securely fastened and return your seat, table, footrest, seat, TV, and controllers to the original position.", "chunks": [ { "text": "Ladies and gentlemen, we are now making our final approach. Please make sure that your seatbelt is securely fastened and return your seat, table, footrest, seat, TV, and controllers to the original position.", "start": 0.0, "end": 14.6, "chunk_index": 0, "language": "English", "finish_reason": "eos", "truncated": false, "generated_tokens": 45, "max_new_tokens": 208 } ], "detected_language": "English", "segments": [ { "start": 1107.3, "end": 1121.8999999999999, "text": "Ladies and gentlemen, we are now making our final approach. Please make sure that your seatbelt is securely fastened and return your seat, table, footrest, seat, TV, and controllers to the original position." } ], "start": 1107.3, "end": 1121.8999999999999, "engine": "mlx-qwen3-asr", "language": "en" }, "raw_word_timestamps": [ { "word": " Please", "start": 1111.6399999999999, "end": 1112.2, "probability": 0.9921875 }, { "word": " make", "start": 1112.2, "end": 1112.48, "probability": 0.97900390625 }, { "word": " sure", "start": 1112.48, "end": 1112.76, "probability": 0.998046875 }, { "word": " that", "start": 1112.76, "end": 1113.06, "probability": 0.98095703125 }, { "word": " your", "start": 1113.06, "end": 1113.28, "probability": 0.935546875 }, { "word": " seat", "start": 1113.28, "end": 1113.58, "probability": 0.98974609375 }, { "word": " belt", "start": 1113.58, "end": 1113.84, "probability": 0.91162109375 }, { "word": " is", "start": 1113.84, "end": 1114.04, "probability": 0.9970703125 }, { "word": " securely", "start": 1114.04, "end": 1114.44, "probability": 0.98095703125 }, { "word": " fastened", "start": 1114.44, "end": 1115.3, "probability": 0.830078125 }, { "word": " and", "start": 1115.3, "end": 1115.52, "probability": 0.75146484375 }, { "word": " will", "start": 1115.52, "end": 1115.6599999999999, "probability": 0.75537109375 }, { "word": " turn", "start": 1115.6599999999999, "end": 1115.86, "probability": 0.92919921875 }, { "word": " your", "start": 1115.86, "end": 1116.02, "probability": 0.96875 }, { "word": " seat", "start": 1116.02, "end": 1116.1599999999999, "probability": 0.994140625 }, { "word": " on", "start": 1116.1599999999999, "end": 1116.3999999999999, "probability": 0.2003173828125 }, { "word": " table,", "start": 1116.3999999999999, "end": 1116.6599999999999, "probability": 0.41162109375 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "Please make sure that your seat belt is securely fastened and return your seat, table,", "review_status": "CERTAIN_RESTORATION", "classification": "CERTAIN_RESTORATION", "confidence": 0.95, "correction_reason": "Source 'will turn' is incorrect; independent ASR confirms 'return' and lists items correctly." }, { "start": 1117.04, "end": 1120.62, "text": "footrest, seat TV and controllers to the original position.", "raw_asr_text": "footless, seat TV and controllers to the original position.", "source_language": "en", "source_lang": "en", "asr_packet_id": 73, "words": [], "asr_metrics": { "avg_logprob": -0.27808494567871095, "compression_ratio": 1.4256756756756757, "no_speech_prob": 6.46698389283884e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "en", "text": "Ladies and gentlemen, we are now making our final approach. Please make sure that your seat belt is securely fastened and will turn your seat on table, footless, seat TV and controllers to the original position." } ], "id": 180, "review_annotation": "", "row_id": 180, "bounded_row_id": 180, "parent_row_ids": [ 180 ], "parent_row_id": 180, "packet_id": 73, "LID_evidence": [ { "region_id": 166, "probabilities": { "ja": 0.00820494735671673, "zh": 0.01211459712976648, "en": 0.9185813516378403 }, "status": "STABLE" }, { "region_id": 167, "probabilities": { "ja": 0.009198353559310948, "zh": 0.011230481921562127, "en": 0.8600826007979256 }, "status": "STABLE" } ], "VAD_region_ids": [ 166, 167 ], "language_confidence": 0.9185813516378403, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1107.3, "packet_end": 1121.8999999999999, "raw_text": "footless, seat TV and controllers to the original position.", "selected_text": "footrest, seat TV and controllers to the original position.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 180 ], "packet_id": 73, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "footless, seat TV and controllers to the original position.", "after": "footrest, seat TV and controllers to the original position.", "classification": "CERTAIN_RESTORATION", "reason": "Source 'footless' is an ASR error for 'footrest', confirmed by independent ASR." } ], "language": "en", "packet_ids": [ 73 ], "index": 180, "independent_ASR": { "text": "Ladies and gentlemen, we are now making our final approach. Please make sure that your seatbelt is securely fastened and return your seat, table, footrest, seat, TV, and controllers to the original position.", "chunks": [ { "text": "Ladies and gentlemen, we are now making our final approach. Please make sure that your seatbelt is securely fastened and return your seat, table, footrest, seat, TV, and controllers to the original position.", "start": 0.0, "end": 14.6, "chunk_index": 0, "language": "English", "finish_reason": "eos", "truncated": false, "generated_tokens": 45, "max_new_tokens": 208 } ], "detected_language": "English", "segments": [ { "start": 1107.3, "end": 1121.8999999999999, "text": "Ladies and gentlemen, we are now making our final approach. Please make sure that your seatbelt is securely fastened and return your seat, table, footrest, seat, TV, and controllers to the original position." } ], "start": 1107.3, "end": 1121.8999999999999, "engine": "mlx-qwen3-asr", "language": "en" }, "raw_word_timestamps": [ { "word": " footless,", "start": 1117.04, "end": 1117.46, "probability": 0.77685546875 }, { "word": " seat", "start": 1117.96, "end": 1118.22, "probability": 0.93017578125 }, { "word": " TV", "start": 1118.22, "end": 1118.48, "probability": 0.87548828125 }, { "word": " and", "start": 1118.48, "end": 1118.7, "probability": 0.7236328125 }, { "word": " controllers", "start": 1118.7, "end": 1119.28, "probability": 0.93017578125 }, { "word": " to", "start": 1119.28, "end": 1119.6399999999999, "probability": 0.994140625 }, { "word": " the", "start": 1119.6399999999999, "end": 1119.8, "probability": 0.96240234375 }, { "word": " original", "start": 1119.8, "end": 1120.1, "probability": 0.80908203125 }, { "word": " position.", "start": 1120.1, "end": 1120.62, "probability": 0.994140625 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "footrest, seat TV and controllers to the original position.", "review_status": "CERTAIN_RESTORATION", "classification": "CERTAIN_RESTORATION", "confidence": 0.95, "correction_reason": "Source 'footless' is an ASR error for 'footrest', confirmed by independent ASR." }, { "start": 1123.5, "end": 1127.1000000000001, "text": "Please refrain from using the lavatories. Thank you.", "raw_asr_text": "Please refrain from using the lavatories. Thank you.", "source_language": "en", "source_lang": "en", "asr_packet_id": 74, "words": [], "asr_metrics": { "avg_logprob": -0.5506945337568011, "compression_ratio": 0.8666666666666667, "no_speech_prob": 5.6159656786070045e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "en", "text": "Please refrain from using the lavatories. Thank you." }, { "language": "ja", "text": "ご視聴ありがとうございました。" } ], "id": 181, "review_annotation": "", "row_id": 181, "bounded_row_id": 181, "parent_row_ids": [ 181 ], "parent_row_id": 181, "packet_id": 74, "LID_evidence": [ { "region_id": 168, "probabilities": { "ja": 0.025814487878233195, "zh": 0.004405851289629936, "en": 0.9138293743133545 }, "status": "STABLE" }, { "region_id": 169, "probabilities": { "ja": 0.013446471886709332, "zh": 0.00296474841888994, "en": 0.9215492606163025 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 168, 169 ], "language_confidence": 0.9215492606163025, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1122.7, "packet_end": 1128.3, "raw_text": "Please refrain from using the lavatories. Thank you.", "selected_text": "Please refrain from using the lavatories. Thank you.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 181 ], "packet_id": 74, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "Please refrain from using the lavatories. Thank you.", "after": "Please refrain from using the lavatories. Thank you.", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Text matches standard safety announcement phrasing and ASR output perfectly." } ], "language": "en", "packet_ids": [ 74 ], "index": 181, "source_text": "Please refrain from using the lavatories. Thank you.", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Text matches standard safety announcement phrasing and ASR output perfectly.", "raw_word_timestamps": [ { "word": " Please", "start": 1122.7, "end": 1123.8600000000001, "probability": 0.85107421875 }, { "word": " refrain", "start": 1123.8600000000001, "end": 1124.14, "probability": 0.191162109375 }, { "word": " from", "start": 1124.14, "end": 1124.44, "probability": 0.99462890625 }, { "word": " using", "start": 1124.44, "end": 1124.8400000000001, "probability": 0.99169921875 }, { "word": " the", "start": 1124.8400000000001, "end": 1125.06, "probability": 0.966796875 }, { "word": " lavatories.", "start": 1125.06, "end": 1125.6000000000001, "probability": 0.82568359375 }, { "word": " Thank", "start": 1126.24, "end": 1126.88, "probability": 0.986328125 }, { "word": " you.", "start": 1126.88, "end": 1127.1000000000001, "probability": 0.96337890625 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 1131.1, "end": 1133.6, "text": "各位贵宾,飞机现在准备降落", "raw_asr_text": "各位贵宾,飞机现在准备降落", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 75, "words": [], "asr_metrics": { "avg_logprob": -0.16634928926508477, "compression_ratio": 1.1452513966480447, "no_speech_prob": 4.356669539018476e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "各位贵宾,飞机现在准备降落请您确实系好安全带请将预备数值餐桌、脚踏板、座位电视、遥控器等受惠原位手提行李请放在前方座位的下面或上方的行李架" } ], "id": 182, "review_annotation": "", "row_id": 182, "bounded_row_id": 182, "parent_row_ids": [ 182 ], "parent_row_id": 182, "packet_id": 75, "LID_evidence": [ { "region_id": 170, "probabilities": { "ja": 0.01547975584187291, "zh": 0.8391416201537306, "en": 0.06522992727431384 }, "status": "STABLE" } ], "VAD_region_ids": [ 170 ], "language_confidence": 0.8391416201537306, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1130.3, "packet_end": 1148.0, "raw_text": "各位贵宾,飞机现在准备降落", "selected_text": "各位贵宾,飞机现在准备降落", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 182 ], "packet_id": 75, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "各位贵宾,飞机现在准备降落", "after": "各位贵宾,飞机现在准备降落", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard announcement phrasing matches ASR and context of final approach." } ], "language": "zh", "packet_ids": [ 75 ], "index": 182, "source_text": "各位贵宾,飞机现在准备降落", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Standard announcement phrasing matches ASR and context of final approach.", "raw_word_timestamps": [ { "word": "各位", "start": 1131.0, "end": 1131.3999999999999, "probability": 0.98828125 }, { "word": "贵", "start": 1131.3999999999999, "end": 1131.62, "probability": 0.68359375 }, { "word": "宾,", "start": 1131.62, "end": 1131.82, "probability": 0.7236328125 }, { "word": "飞", "start": 1132.04, "end": 1132.22, "probability": 0.951171875 }, { "word": "机", "start": 1132.22, "end": 1132.36, "probability": 0.986328125 }, { "word": "现在", "start": 1132.36, "end": 1132.6399999999999, "probability": 0.98193359375 }, { "word": "准", "start": 1132.6399999999999, "end": 1132.94, "probability": 0.99462890625 }, { "word": "备", "start": 1132.94, "end": 1133.1, "probability": 1.0 }, { "word": "降", "start": 1133.1, "end": 1133.32, "probability": 0.99951171875 }, { "word": "落", "start": 1133.32, "end": 1133.6, "probability": 0.998046875 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 1133.6, "end": 1135.9199999999998, "text": "请您确实系好安全带", "raw_asr_text": "请您确实系好安全带", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 75, "words": [ { "word": "请", "start": 1133.6, "end": 1134.34, "probability": 0.58447265625 }, { "word": "您", "start": 1134.34, "end": 1134.52, "probability": 0.93994140625 }, { "word": "确", "start": 1134.52, "end": 1134.68, "probability": 0.990234375 }, { "word": "实", "start": 1134.68, "end": 1134.86, "probability": 0.99755859375 }, { "word": "系", "start": 1134.86, "end": 1135.0, "probability": 0.533203125 }, { "word": "好", "start": 1135.0, "end": 1135.2, "probability": 0.912109375 }, { "word": "安全", "start": 1135.2, "end": 1135.48, "probability": 0.9755859375 }, { "word": "带", "start": 1135.48, "end": 1135.9199999999998, "probability": 0.97509765625 } ], "asr_metrics": { "avg_logprob": -0.16634928926508477, "compression_ratio": 1.1452513966480447, "no_speech_prob": 4.356669539018476e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "各位贵宾,飞机现在准备降落请您确实系好安全带请将预备数值餐桌、脚踏板、座位电视、遥控器等受惠原位手提行李请放在前方座位的下面或上方的行李架" } ], "id": 183, "review_annotation": "", "row_id": 183, "bounded_row_id": 183, "parent_row_ids": [ 183 ], "parent_row_id": 183, "packet_id": 75, "LID_evidence": [ { "region_id": 170, "probabilities": { "ja": 0.01547975584187291, "zh": 0.8391416201537306, "en": 0.06522992727431384 }, "status": "STABLE" } ], "VAD_region_ids": [ 170 ], "language_confidence": 0.8391416201537306, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1130.3, "packet_end": 1148.0, "raw_text": "请您确实系好安全带", "selected_text": "请您确实系好安全带", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 183 ], "packet_id": 75, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "请您确实系好安全带", "after": "请您确实系好安全带", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard safety instruction matches ASR and context." } ], "language": "zh", "packet_ids": [ 75 ], "index": 183, "source_text": "请您确实系好安全带", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Standard safety instruction matches ASR and context." }, { "start": 1136.8799999999999, "end": 1138.24, "text": "請將椅背豎直", "raw_asr_text": "请将预备数值", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 75, "words": [], "asr_metrics": { "avg_logprob": -0.16634928926508477, "compression_ratio": 1.1452513966480447, "no_speech_prob": 4.356669539018476e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "各位贵宾,飞机现在准备降落请您确实系好安全带请将预备数值餐桌、脚踏板、座位电视、遥控器等受惠原位手提行李请放在前方座位的下面或上方的行李架" } ], "id": 184, "review_annotation": "", "row_id": 184, "bounded_row_id": 184, "parent_row_ids": [ 184 ], "parent_row_id": 184, "packet_id": 75, "LID_evidence": [ { "region_id": 171, "probabilities": { "ja": 0.007230623680654736, "zh": 0.9231283845646041, "en": 0.029464720582057322 }, "status": "STABLE" } ], "VAD_region_ids": [ 171 ], "language_confidence": 0.9231283845646041, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1130.3, "packet_end": 1148.0, "raw_text": "请将预备数值", "selected_text": "請將椅背豎直", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 184 ], "packet_id": 75, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "请将预备数值", "after": "請將椅背豎直", "classification": "CERTAIN_RESTORATION", "reason": "Source '預備數值' is nonsense; independent ASR confirms 'upright seat back' instruction." } ], "language": "zh", "packet_ids": [ 75 ], "index": 184, "independent_ASR": { "text": "各位贵宾，飞机现在准备降落，请您确实系好安全带，请将椅背竖直，餐桌、脚踏板、座位电视、遥控器等收回原位，手提行李请放在前方座位的下面或上方的行李架。", "chunks": [ { "text": "各位贵宾，飞机现在准备降落，请您确实系好安全带，请将椅背竖直，餐桌、脚踏板、座位电视、遥控器等收回原位，手提行李请放在前方座位的下面或上方的行李架。", "start": 0.0, "end": 17.7, "chunk_index": 0, "language": "Chinese", "finish_reason": "eos", "truncated": false, "generated_tokens": 52, "max_new_tokens": 245 } ], "detected_language": "Chinese", "segments": [ { "start": 1130.3, "end": 1148.0, "text": "各位贵宾，飞机现在准备降落，请您确实系好安全带，请将椅背竖直，餐桌、脚踏板、座位电视、遥控器等收回原位，手提行李请放在前方座位的下面或上方的行李架。" } ], "start": 1130.3, "end": 1148.0, "engine": "mlx-qwen3-asr", "language": "zh" }, "raw_word_timestamps": [ { "word": "请", "start": 1136.8799999999999, "end": 1137.28, "probability": 0.97021484375 }, { "word": "将", "start": 1137.28, "end": 1137.48, "probability": 0.9912109375 }, { "word": "预", "start": 1137.48, "end": 1137.62, "probability": 0.462890625 }, { "word": "备", "start": 1137.62, "end": 1137.8, "probability": 0.9990234375 }, { "word": "数", "start": 1137.8, "end": 1137.94, "probability": 0.468017578125 }, { "word": "值", "start": 1137.94, "end": 1138.24, "probability": 0.931640625 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "請將椅背豎直", "review_status": "CERTAIN_RESTORATION", "classification": "CERTAIN_RESTORATION", "confidence": 0.95, "correction_reason": "Source '預備數值' is nonsense; independent ASR confirms 'upright seat back' instruction." }, { "start": 1138.24, "end": 1142.58, "text": "餐桌、腳踏板、座位電視、遙控器等收回原位", "raw_asr_text": "餐桌、脚踏板、座位电视、遥控器等受惠原位", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 75, "words": [], "asr_metrics": { "avg_logprob": -0.16634928926508477, "compression_ratio": 1.1452513966480447, "no_speech_prob": 4.356669539018476e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "各位贵宾,飞机现在准备降落请您确实系好安全带请将预备数值餐桌、脚踏板、座位电视、遥控器等受惠原位手提行李请放在前方座位的下面或上方的行李架" } ], "id": 185, "review_annotation": "", "row_id": 185, "bounded_row_id": 185, "parent_row_ids": [ 185 ], "parent_row_id": 185, "packet_id": 75, "LID_evidence": [ { "region_id": 171, "probabilities": { "ja": 0.007230623680654736, "zh": 0.9231283845646041, "en": 0.029464720582057322 }, "status": "STABLE" } ], "VAD_region_ids": [ 171 ], "language_confidence": 0.9231283845646041, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1130.3, "packet_end": 1148.0, "raw_text": "餐桌、脚踏板、座位电视、遥控器等受惠原位", "selected_text": "餐桌、腳踏板、座位電視、遙控器等收回原位", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 185 ], "packet_id": 75, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "餐桌、脚踏板、座位电视、遥控器等受惠原位", "after": "餐桌、腳踏板、座位電視、遙控器等收回原位", "classification": "CERTAIN_RESTORATION", "reason": "Source '受惠' is a phonetic error for '收回' (return/stow), confirmed by independent ASR." } ], "language": "zh", "packet_ids": [ 75 ], "index": 185, "independent_ASR": { "text": "各位贵宾，飞机现在准备降落，请您确实系好安全带，请将椅背竖直，餐桌、脚踏板、座位电视、遥控器等收回原位，手提行李请放在前方座位的下面或上方的行李架。", "chunks": [ { "text": "各位贵宾，飞机现在准备降落，请您确实系好安全带，请将椅背竖直，餐桌、脚踏板、座位电视、遥控器等收回原位，手提行李请放在前方座位的下面或上方的行李架。", "start": 0.0, "end": 17.7, "chunk_index": 0, "language": "Chinese", "finish_reason": "eos", "truncated": false, "generated_tokens": 52, "max_new_tokens": 245 } ], "detected_language": "Chinese", "segments": [ { "start": 1130.3, "end": 1148.0, "text": "各位贵宾，飞机现在准备降落，请您确实系好安全带，请将椅背竖直，餐桌、脚踏板、座位电视、遥控器等收回原位，手提行李请放在前方座位的下面或上方的行李架。" } ], "start": 1130.3, "end": 1148.0, "engine": "mlx-qwen3-asr", "language": "zh" }, "raw_word_timestamps": [ { "word": "餐", "start": 1138.24, "end": 1138.98, "probability": 0.880859375 }, { "word": "桌、", "start": 1138.98, "end": 1139.22, "probability": 1.0 }, { "word": "脚", "start": 1139.34, "end": 1139.46, "probability": 0.99169921875 }, { "word": "踏", "start": 1139.46, "end": 1139.6, "probability": 0.998046875 }, { "word": "板、", "start": 1139.6, "end": 1139.84, "probability": 0.99951171875 }, { "word": "座", "start": 1140.0, "end": 1140.12, "probability": 0.91748046875 }, { "word": "位", "start": 1140.12, "end": 1140.24, "probability": 0.99951171875 }, { "word": "电", "start": 1140.24, "end": 1140.44, "probability": 0.9912109375 }, { "word": "视、", "start": 1140.44, "end": 1140.72, "probability": 0.99560546875 }, { "word": "遥", "start": 1140.86, "end": 1141.0, "probability": 0.97314453125 }, { "word": "控", "start": 1141.0, "end": 1141.2, "probability": 0.998046875 }, { "word": "器", "start": 1141.2, "end": 1141.4199999999998, "probability": 0.99951171875 }, { "word": "等", "start": 1141.4199999999998, "end": 1141.6599999999999, "probability": 0.9970703125 }, { "word": "受", "start": 1141.6599999999999, "end": 1141.94, "probability": 0.283447265625 }, { "word": "惠", "start": 1141.94, "end": 1142.12, "probability": 0.70751953125 }, { "word": "原", "start": 1142.12, "end": 1142.28, "probability": 0.92431640625 }, { "word": "位", "start": 1142.28, "end": 1142.58, "probability": 0.94482421875 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "餐桌、腳踏板、座位電視、遙控器等收回原位", "review_status": "CERTAIN_RESTORATION", "classification": "CERTAIN_RESTORATION", "confidence": 0.95, "correction_reason": "Source '受惠' is a phonetic error for '收回' (return/stow), confirmed by independent ASR." }, { "start": 1142.58, "end": 1146.9199999999998, "text": "手提行李请放在前方座位的下面或上方的行李架", "raw_asr_text": "手提行李请放在前方座位的下面或上方的行李架", "source_language": "zh", "source_lang": "zh", "asr_packet_id": 75, "words": [ { "word": "手", "start": 1142.58, "end": 1143.32, "probability": 0.83251953125 }, { "word": "提", "start": 1143.32, "end": 1143.48, "probability": 0.98828125 }, { "word": "行", "start": 1143.48, "end": 1143.6599999999999, "probability": 0.88916015625 }, { "word": "李", "start": 1143.6599999999999, "end": 1143.82, "probability": 0.99462890625 }, { "word": "请", "start": 1143.82, "end": 1144.02, "probability": 0.90234375 }, { "word": "放", "start": 1144.02, "end": 1144.24, "probability": 0.994140625 }, { "word": "在", "start": 1144.24, "end": 1144.46, "probability": 0.99951171875 }, { "word": "前", "start": 1144.46, "end": 1144.6399999999999, "probability": 0.99560546875 }, { "word": "方", "start": 1144.6399999999999, "end": 1144.86, "probability": 0.99609375 }, { "word": "座", "start": 1144.86, "end": 1145.04, "probability": 1.0 }, { "word": "位", "start": 1145.04, "end": 1145.18, "probability": 1.0 }, { "word": "的", "start": 1145.18, "end": 1145.3, "probability": 0.986328125 }, { "word": "下面", "start": 1145.3, "end": 1145.6, "probability": 0.99951171875 }, { "word": "或", "start": 1145.6, "end": 1145.8799999999999, "probability": 0.9970703125 }, { "word": "上", "start": 1145.8799999999999, "end": 1146.1, "probability": 0.9580078125 }, { "word": "方", "start": 1146.1, "end": 1146.26, "probability": 0.99951171875 }, { "word": "的", "start": 1146.26, "end": 1146.4199999999998, "probability": 0.9990234375 }, { "word": "行", "start": 1146.4199999999998, "end": 1146.56, "probability": 0.98486328125 }, { "word": "李", "start": 1146.56, "end": 1146.74, "probability": 0.9990234375 }, { "word": "架", "start": 1146.74, "end": 1146.9199999999998, "probability": 0.7392578125 } ], "asr_metrics": { "avg_logprob": -0.16634928926508477, "compression_ratio": 1.1452513966480447, "no_speech_prob": 4.356669539018476e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "zh", "text": "各位贵宾,飞机现在准备降落请您确实系好安全带请将预备数值餐桌、脚踏板、座位电视、遥控器等受惠原位手提行李请放在前方座位的下面或上方的行李架" } ], "id": 186, "review_annotation": "", "row_id": 186, "bounded_row_id": 186, "parent_row_ids": [ 186 ], "parent_row_id": 186, "packet_id": 75, "LID_evidence": [ { "region_id": 171, "probabilities": { "ja": 0.007230623680654736, "zh": 0.9231283845646041, "en": 0.029464720582057322 }, "status": "STABLE" } ], "VAD_region_ids": [ 171 ], "language_confidence": 0.9231283845646041, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1130.3, "packet_end": 1148.0, "raw_text": "手提行李请放在前方座位的下面或上方的行李架", "selected_text": "手提行李请放在前方座位的下面或上方的行李架", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 186 ], "packet_id": 75, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "手提行李请放在前方座位的下面或上方的行李架", "after": "手提行李请放在前方座位的下面或上方的行李架", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard instruction matches ASR and context perfectly." } ], "language": "zh", "packet_ids": [ 75 ], "index": 186, "source_text": "手提行李请放在前方座位的下面或上方的行李架", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Standard instruction matches ASR and context perfectly." }, { "start": 1151.0, "end": 1154.0, "text": "皆様、羽田空港に着陸いたしました。", "raw_asr_text": "皆様、羽田空港に着陸いたしました。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 76, "words": [], "asr_metrics": { "avg_logprob": -0.20815523465474448, "compression_ratio": 0.85, "no_speech_prob": 3.773403464690794e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "皆様、羽田空港に着陸いたしました。" } ], "id": 187, "review_annotation": "", "row_id": 187, "bounded_row_id": 187, "parent_row_ids": [ 187 ], "parent_row_id": 187, "packet_id": 76, "LID_evidence": [ { "region_id": 172, "probabilities": { "ja": 0.985502228140831, "zh": 0.0008773556437517982, "en": 0.006746347149601206 }, "status": "STABLE" } ], "VAD_region_ids": [ 172 ], "language_confidence": 0.985502228140831, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1150.2, "packet_end": 1155.3, "raw_text": "皆様、羽田空港に着陸いたしました。", "selected_text": "皆様、羽田空港に着陸いたしました。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 187 ], "packet_id": 76, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "皆様、羽田空港に着陸いたしました。", "after": "皆様、羽田空港に着陸いたしました。", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard announcement matches ASR and context." } ], "language": "ja", "packet_ids": [ 76 ], "index": 187, "source_text": "皆様、羽田空港に着陸いたしました。", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard announcement matches ASR and context.", "raw_word_timestamps": [ { "word": "皆", "start": 1150.88, "end": 1151.28, "probability": 0.96240234375 }, { "word": "様、", "start": 1151.28, "end": 1151.5800000000002, "probability": 0.60009765625 }, { "word": "羽", "start": 1151.76, "end": 1151.94, "probability": 0.99658203125 }, { "word": "田", "start": 1151.94, "end": 1152.1000000000001, "probability": 0.99755859375 }, { "word": "空", "start": 1152.1000000000001, "end": 1152.3, "probability": 0.99951171875 }, { "word": "港", "start": 1152.3, "end": 1152.6200000000001, "probability": 1.0 }, { "word": "に", "start": 1152.6200000000001, "end": 1152.82, "probability": 0.99853515625 }, { "word": "着", "start": 1152.82, "end": 1153.06, "probability": 0.9990234375 }, { "word": "陸", "start": 1153.06, "end": 1153.38, "probability": 0.99951171875 }, { "word": "いた", "start": 1153.38, "end": 1153.54, "probability": 0.98193359375 }, { "word": "しました。", "start": 1153.54, "end": 1154.0, "probability": 1.0 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 1157.5, "end": 1159.96, "text": "ただいまの時刻は3月14日", "raw_asr_text": "ただいまの時刻は3月14日", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 77, "words": [], "asr_metrics": { "avg_logprob": -0.22793126106262207, "compression_ratio": 0.75, "no_speech_prob": 5.4048358670566454e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "ただいまの時刻は3月14日" } ], "id": 188, "review_annotation": "", "row_id": 188, "bounded_row_id": 188, "parent_row_ids": [ 188 ], "parent_row_id": 188, "packet_id": 77, "LID_evidence": [ { "region_id": 173, "probabilities": { "ja": 0.9953052600224813, "zh": 0.00040731799041774747, "en": 0.001952109334524721 }, "status": "STABLE" } ], "VAD_region_ids": [ 173 ], "language_confidence": 0.9953052600224813, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1156.7, "packet_end": 1161.2, "raw_text": "ただいまの時刻は3月14日", "selected_text": "ただいまの時刻は3月14日", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 188 ], "packet_id": 77, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "ただいまの時刻は3月14日", "after": "ただいまの時刻は3月14日", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard announcement matches ASR and context." } ], "language": "ja", "packet_ids": [ 77 ], "index": 188, "source_text": "ただいまの時刻は3月14日", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard announcement matches ASR and context.", "raw_word_timestamps": [ { "word": "ただ", "start": 1157.24, "end": 1157.72, "probability": 0.96533203125 }, { "word": "い", "start": 1157.72, "end": 1157.82, "probability": 0.974609375 }, { "word": "ま", "start": 1157.82, "end": 1157.92, "probability": 0.99853515625 }, { "word": "の", "start": 1157.92, "end": 1158.0800000000002, "probability": 0.9990234375 }, { "word": "時", "start": 1158.0800000000002, "end": 1158.18, "probability": 0.9990234375 }, { "word": "刻", "start": 1158.18, "end": 1158.4, "probability": 0.99951171875 }, { "word": "は", "start": 1158.4, "end": 1158.64, "probability": 1.0 }, { "word": "3", "start": 1158.64, "end": 1158.94, "probability": 0.7783203125 }, { "word": "月", "start": 1158.94, "end": 1159.18, "probability": 1.0 }, { "word": "14", "start": 1159.18, "end": 1159.6200000000001, "probability": 0.99951171875 }, { "word": "日", "start": 1159.6200000000001, "end": 1159.96, "probability": 1.0 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 1162.6, "end": 1164.7, "text": "午後5時8分でございます", "raw_asr_text": "午後5時8分でございます", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 78, "words": [], "asr_metrics": { "avg_logprob": -0.2924306175925515, "compression_ratio": 0.7441860465116279, "no_speech_prob": 4.976890861235761e-12 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "午後5時8分でございます" } ], "id": 189, "review_annotation": "", "row_id": 189, "bounded_row_id": 189, "parent_row_ids": [ 189 ], "parent_row_id": 189, "packet_id": 78, "LID_evidence": [ { "region_id": 174, "probabilities": { "ja": 0.9949220418930054, "zh": 0.00042787053826032206, "en": 0.0023159555275924504 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 174 ], "language_confidence": 0.9949220418930054, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1161.8, "packet_end": 1165.8999999999999, "raw_text": "午後5時8分でございます", "selected_text": "午後5時8分でございます", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 189 ], "packet_id": 78, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "午後5時8分でございます", "after": "午後5時8分でございます", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard announcement matches ASR and context." } ], "language": "ja", "packet_ids": [ 78 ], "index": 189, "source_text": "午後5時8分でございます", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard announcement matches ASR and context.", "raw_word_timestamps": [ { "word": "午", "start": 1162.34, "end": 1162.84, "probability": 0.978515625 }, { "word": "後", "start": 1162.84, "end": 1163.08, "probability": 0.99951171875 }, { "word": "5", "start": 1163.08, "end": 1163.26, "probability": 0.994140625 }, { "word": "時", "start": 1163.26, "end": 1163.48, "probability": 1.0 }, { "word": "8", "start": 1163.48, "end": 1163.82, "probability": 0.9951171875 }, { "word": "分", "start": 1163.82, "end": 1164.08, "probability": 0.99951171875 }, { "word": "で", "start": 1164.08, "end": 1164.2, "probability": 0.99560546875 }, { "word": "ございます", "start": 1164.2, "end": 1164.7, "probability": 1.0 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 1168.0, "end": 1171.82, "text": "ベルト着用サインが消えるまで、お座りのままお待ちください。", "raw_asr_text": "ベルト着用サインが消えるまで、お座りのままお待ちください。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 79, "words": [], "asr_metrics": { "avg_logprob": -0.13614564282553537, "compression_ratio": 1.0481927710843373, "no_speech_prob": 4.37807168990334e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "ベルト着用サインが消えるまで、お座りのままお待ちください。" } ], "id": 190, "review_annotation": "", "row_id": 190, "bounded_row_id": 190, "parent_row_ids": [ 190 ], "parent_row_id": 190, "packet_id": 79, "LID_evidence": [ { "region_id": 175, "probabilities": { "ja": 0.9963774581750234, "zh": 0.00022439857517989972, "en": 0.0016127265116665512 }, "status": "STABLE" } ], "VAD_region_ids": [ 175 ], "language_confidence": 0.9963774581750234, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1167.2, "packet_end": 1173.1, "raw_text": "ベルト着用サインが消えるまで、お座りのままお待ちください。", "selected_text": "ベルト着用サインが消えるまで、お座りのままお待ちください。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 190 ], "packet_id": 79, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "ベルト着用サインが消えるまで、お座りのままお待ちください。", "after": "ベルト着用サインが消えるまで、お座りのままお待ちください。", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard safety announcement phrasing matches context and raw ASR perfectly." } ], "language": "ja", "packet_ids": [ 79 ], "index": 190, "source_text": "ベルト着用サインが消えるまで、お座りのままお待ちください。", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard safety announcement phrasing matches context and raw ASR perfectly.", "raw_word_timestamps": [ { "word": "ベ", "start": 1167.92, "end": 1168.24, "probability": 0.9912109375 }, { "word": "ル", "start": 1168.24, "end": 1168.3400000000001, "probability": 0.99951171875 }, { "word": "ト", "start": 1168.3400000000001, "end": 1168.44, "probability": 0.99951171875 }, { "word": "着", "start": 1168.44, "end": 1168.6000000000001, "probability": 0.99951171875 }, { "word": "用", "start": 1168.6000000000001, "end": 1168.8600000000001, "probability": 1.0 }, { "word": "サ", "start": 1168.8600000000001, "end": 1169.1000000000001, "probability": 0.98486328125 }, { "word": "イ", "start": 1169.1000000000001, "end": 1169.2, "probability": 1.0 }, { "word": "ン", "start": 1169.2, "end": 1169.28, "probability": 1.0 }, { "word": "が", "start": 1169.28, "end": 1169.38, "probability": 0.99951171875 }, { "word": "消", "start": 1169.38, "end": 1169.56, "probability": 0.9990234375 }, { "word": "える", "start": 1169.56, "end": 1169.74, "probability": 0.99951171875 }, { "word": "まで、", "start": 1169.74, "end": 1170.02, "probability": 0.99951171875 }, { "word": "お", "start": 1170.2, "end": 1170.32, "probability": 0.99853515625 }, { "word": "座", "start": 1170.32, "end": 1170.52, "probability": 0.99853515625 }, { "word": "り", "start": 1170.52, "end": 1170.64, "probability": 1.0 }, { "word": "の", "start": 1170.64, "end": 1170.76, "probability": 1.0 }, { "word": "ま", "start": 1170.76, "end": 1170.9, "probability": 1.0 }, { "word": "ま", "start": 1170.9, "end": 1171.06, "probability": 1.0 }, { "word": "お", "start": 1171.06, "end": 1171.24, "probability": 0.9990234375 }, { "word": "待", "start": 1171.24, "end": 1171.4, "probability": 1.0 }, { "word": "ち", "start": 1171.4, "end": 1171.48, "probability": 0.9990234375 }, { "word": "ください。", "start": 1171.48, "end": 1171.82, "probability": 0.99853515625 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 1174.9800000000002, "end": 1179.8600000000001, "text": "物入れを開けたときに手荷物が滑り出る恐れがありますのでお気を付けください", "raw_asr_text": "物入れを開けたときに手荷物が滑り出る恐れがありますのでお気を付けください", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 80, "words": [ { "word": "物", "start": 1174.9800000000002, "end": 1175.22, "probability": 0.9716796875 }, { "word": "入", "start": 1175.22, "end": 1175.3600000000001, "probability": 0.99951171875 }, { "word": "れ", "start": 1175.3600000000001, "end": 1175.44, "probability": 0.99658203125 }, { "word": "を", "start": 1175.44, "end": 1175.5400000000002, "probability": 0.99951171875 }, { "word": "開", "start": 1175.5400000000002, "end": 1175.66, "probability": 0.9619140625 }, { "word": "け", "start": 1175.66, "end": 1175.7600000000002, "probability": 1.0 }, { "word": "た", "start": 1175.7600000000002, "end": 1175.88, "probability": 1.0 }, { "word": "と", "start": 1175.88, "end": 1176.0000000000002, "probability": 0.60009765625 }, { "word": "き", "start": 1176.0000000000002, "end": 1176.1200000000001, "probability": 1.0 }, { "word": "に", "start": 1176.1200000000001, "end": 1176.3400000000001, "probability": 1.0 }, { "word": "手", "start": 1176.3400000000001, "end": 1176.5800000000002, "probability": 0.43115234375 }, { "word": "荷", "start": 1176.5800000000002, "end": 1176.7, "probability": 0.9951171875 }, { "word": "物", "start": 1176.7, "end": 1176.92, "probability": 1.0 }, { "word": "が", "start": 1176.92, "end": 1177.14, "probability": 0.99951171875 }, { "word": "滑", "start": 1177.14, "end": 1177.46, "probability": 0.99755859375 }, { "word": "り", "start": 1177.46, "end": 1177.5400000000002, "probability": 1.0 }, { "word": "出", "start": 1177.5400000000002, "end": 1177.66, "probability": 0.99609375 }, { "word": "る", "start": 1177.66, "end": 1177.7600000000002, "probability": 0.99951171875 }, { "word": "恐", "start": 1177.7600000000002, "end": 1177.94, "probability": 0.64208984375 }, { "word": "れ", "start": 1177.94, "end": 1178.0800000000002, "probability": 1.0 }, { "word": "が", "start": 1178.0800000000002, "end": 1178.18, "probability": 1.0 }, { "word": "あります", "start": 1178.18, "end": 1178.5400000000002, "probability": 0.998046875 }, { "word": "ので", "start": 1178.5400000000002, "end": 1178.8400000000001, "probability": 1.0 }, { "word": "お", "start": 1178.8400000000001, "end": 1179.0400000000002, "probability": 0.498779296875 }, { "word": "気", "start": 1179.0400000000002, "end": 1179.16, "probability": 1.0 }, { "word": "を", "start": 1179.16, "end": 1179.2600000000002, "probability": 0.99951171875 }, { "word": "付", "start": 1179.2600000000002, "end": 1179.3400000000001, "probability": 0.499755859375 }, { "word": "け", "start": 1179.3400000000001, "end": 1179.46, "probability": 0.99951171875 }, { "word": "ください", "start": 1179.46, "end": 1179.8600000000001, "probability": 0.99560546875 } ], "asr_metrics": { "avg_logprob": -0.16840734201319077, "compression_ratio": 1.0588235294117647, "no_speech_prob": 4.960874159382067e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "物入れを開けたときに手荷物が滑り出る恐れがありますのでお気を付けください" } ], "id": 191, "review_annotation": "", "row_id": 191, "bounded_row_id": 191, "parent_row_ids": [ 191 ], "parent_row_id": 191, "packet_id": 80, "LID_evidence": [ { "region_id": 176, "probabilities": { "ja": 0.9924393221735954, "zh": 0.0003770397361222422, "en": 0.0034263131201441865 }, "status": "STABLE" } ], "VAD_region_ids": [ 176 ], "language_confidence": 0.9924393221735954, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1174.1000000000001, "packet_end": 1181.2, "raw_text": "物入れを開けたときに手荷物が滑り出る恐れがありますのでお気を付けください", "selected_text": "物入れを開けたときに手荷物が滑り出る恐れがありますのでお気を付けください", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 191 ], "packet_id": 80, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "物入れを開けたときに手荷物が滑り出る恐れがありますのでお気を付けください", "after": "物入れを開けたときに手荷物が滑り出る恐れがありますのでお気を付けください", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard safety announcement phrasing matches context and raw ASR perfectly." } ], "language": "ja", "packet_ids": [ 80 ], "index": 191, "source_text": "物入れを開けたときに手荷物が滑り出る恐れがありますのでお気を付けください", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard safety announcement phrasing matches context and raw ASR perfectly." }, { "start": 1184.7, "end": 1189.02, "text": "この先、通信用の電波を発する電子機器もご使用になれますが、", "raw_asr_text": "この先、通信用の電波を発する電子機器もご使用になれますが、", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 81, "words": [], "asr_metrics": { "avg_logprob": -0.1204914042824193, "compression_ratio": 1.3938053097345133, "no_speech_prob": 5.343340787195006e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "この先、通信用の電波を発する電子機器もご使用になれますが、携帯電話での通話は周りのお客様のご迷惑になりますので、お控えください。果物、野菜、植物、肉類は、税関検査の前に、検疫同植物カウンターでの検査が必要です。" } ], "id": 192, "review_annotation": "", "row_id": 192, "bounded_row_id": 192, "parent_row_ids": [ 192 ], "parent_row_id": 192, "packet_id": 81, "LID_evidence": [ { "region_id": 177, "probabilities": { "ja": 0.9952520903419045, "zh": 0.00037314772506600577, "en": 0.0021109394129940912 }, "status": "STABLE" } ], "VAD_region_ids": [ 177 ], "language_confidence": 0.9952520903419045, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1183.9, "packet_end": 1205.5, "raw_text": "この先、通信用の電波を発する電子機器もご使用になれますが、", "selected_text": "この先、通信用の電波を発する電子機器もご使用になれますが、", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 192 ], "packet_id": 81, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "この先、通信用の電波を発する電子機器もご使用になれますが、", "after": "この先、通信用の電波を発する電子機器もご使用になれますが、", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard announcement phrasing matches context and raw ASR perfectly." } ], "language": "ja", "packet_ids": [ 81 ], "index": 192, "source_text": "この先、通信用の電波を発する電子機器もご使用になれますが、", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard announcement phrasing matches context and raw ASR perfectly.", "raw_word_timestamps": [ { "word": "この", "start": 1184.64, "end": 1185.0, "probability": 0.99609375 }, { "word": "先、", "start": 1185.0, "end": 1185.3000000000002, "probability": 0.99951171875 }, { "word": "通", "start": 1185.48, "end": 1185.6200000000001, "probability": 1.0 }, { "word": "信", "start": 1185.6200000000001, "end": 1185.8200000000002, "probability": 0.99951171875 }, { "word": "用", "start": 1185.8200000000002, "end": 1185.98, "probability": 1.0 }, { "word": "の", "start": 1185.98, "end": 1186.14, "probability": 0.9990234375 }, { "word": "電", "start": 1186.14, "end": 1186.26, "probability": 1.0 }, { "word": "波", "start": 1186.26, "end": 1186.44, "probability": 0.97998046875 }, { "word": "を", "start": 1186.44, "end": 1186.6000000000001, "probability": 0.99951171875 }, { "word": "発", "start": 1186.6000000000001, "end": 1186.8400000000001, "probability": 0.98583984375 }, { "word": "する", "start": 1186.8400000000001, "end": 1187.14, "probability": 0.9736328125 }, { "word": "電", "start": 1187.14, "end": 1187.3400000000001, "probability": 0.998046875 }, { "word": "子", "start": 1187.3400000000001, "end": 1187.48, "probability": 0.9951171875 }, { "word": "機", "start": 1187.48, "end": 1187.6000000000001, "probability": 0.998046875 }, { "word": "器", "start": 1187.6000000000001, "end": 1187.74, "probability": 0.99755859375 }, { "word": "も", "start": 1187.74, "end": 1187.88, "probability": 0.99951171875 }, { "word": "ご", "start": 1187.88, "end": 1188.0400000000002, "probability": 0.982421875 }, { "word": "使", "start": 1188.0400000000002, "end": 1188.18, "probability": 0.99853515625 }, { "word": "用", "start": 1188.18, "end": 1188.3000000000002, "probability": 1.0 }, { "word": "にな", "start": 1188.3000000000002, "end": 1188.46, "probability": 0.990234375 }, { "word": "れ", "start": 1188.46, "end": 1188.6000000000001, "probability": 0.986328125 }, { "word": "ます", "start": 1188.6000000000001, "end": 1188.8400000000001, "probability": 1.0 }, { "word": "が、", "start": 1188.8400000000001, "end": 1189.02, "probability": 1.0 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 1189.44, "end": 1194.14, "text": "携帯電話での通話は周りのお客様のご迷惑になりますので、お控えください。", "raw_asr_text": "携帯電話での通話は周りのお客様のご迷惑になりますので、お控えください。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 81, "words": [ { "word": "携", "start": 1189.44, "end": 1189.64, "probability": 0.9921875 }, { "word": "帯", "start": 1189.64, "end": 1189.8000000000002, "probability": 0.99951171875 }, { "word": "電", "start": 1189.8000000000002, "end": 1189.98, "probability": 1.0 }, { "word": "話", "start": 1189.98, "end": 1190.1200000000001, "probability": 1.0 }, { "word": "で", "start": 1190.1200000000001, "end": 1190.22, "probability": 0.99951171875 }, { "word": "の", "start": 1190.22, "end": 1190.3600000000001, "probability": 1.0 }, { "word": "通", "start": 1190.3600000000001, "end": 1190.5600000000002, "probability": 0.99951171875 }, { "word": "話", "start": 1190.5600000000002, "end": 1190.72, "probability": 0.9990234375 }, { "word": "は", "start": 1190.72, "end": 1190.9, "probability": 0.9990234375 }, { "word": "周", "start": 1190.9, "end": 1191.1000000000001, "probability": 0.869140625 }, { "word": "り", "start": 1191.1000000000001, "end": 1191.24, "probability": 1.0 }, { "word": "の", "start": 1191.24, "end": 1191.3600000000001, "probability": 1.0 }, { "word": "お", "start": 1191.3600000000001, "end": 1191.44, "probability": 1.0 }, { "word": "客", "start": 1191.44, "end": 1191.5600000000002, "probability": 1.0 }, { "word": "様", "start": 1191.5600000000002, "end": 1191.8000000000002, "probability": 0.9892578125 }, { "word": "の", "start": 1191.8000000000002, "end": 1191.96, "probability": 0.99951171875 }, { "word": "ご", "start": 1191.96, "end": 1192.1000000000001, "probability": 0.99853515625 }, { "word": "迷", "start": 1192.1000000000001, "end": 1192.24, "probability": 1.0 }, { "word": "惑", "start": 1192.24, "end": 1192.44, "probability": 1.0 }, { "word": "にな", "start": 1192.44, "end": 1192.66, "probability": 1.0 }, { "word": "ります", "start": 1192.66, "end": 1193.0, "probability": 0.99951171875 }, { "word": "ので、", "start": 1193.0, "end": 1193.26, "probability": 1.0 }, { "word": "お", "start": 1193.42, "end": 1193.44, "probability": 0.99609375 }, { "word": "控", "start": 1193.44, "end": 1193.6000000000001, "probability": 0.99951171875 }, { "word": "え", "start": 1193.6000000000001, "end": 1193.72, "probability": 1.0 }, { "word": "ください。", "start": 1193.72, "end": 1194.14, "probability": 0.99951171875 } ], "asr_metrics": { "avg_logprob": -0.1204914042824193, "compression_ratio": 1.3938053097345133, "no_speech_prob": 5.343340787195006e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "この先、通信用の電波を発する電子機器もご使用になれますが、携帯電話での通話は周りのお客様のご迷惑になりますので、お控えください。果物、野菜、植物、肉類は、税関検査の前に、検疫同植物カウンターでの検査が必要です。" } ], "id": 193, "review_annotation": "", "row_id": 193, "bounded_row_id": 193, "parent_row_ids": [ 193 ], "parent_row_id": 193, "packet_id": 81, "LID_evidence": [ { "region_id": 177, "probabilities": { "ja": 0.9952520903419045, "zh": 0.00037314772506600577, "en": 0.0021109394129940912 }, "status": "STABLE" } ], "VAD_region_ids": [ 177 ], "language_confidence": 0.9952520903419045, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1183.9, "packet_end": 1205.5, "raw_text": "携帯電話での通話は周りのお客様のご迷惑になりますので、お控えください。", "selected_text": "携帯電話での通話は周りのお客様のご迷惑になりますので、お控えください。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 193 ], "packet_id": 81, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "携帯電話での通話は周りのお客様のご迷惑になりますので、お控えください。", "after": "携帯電話での通話は周りのお客様のご迷惑になりますので、お控えください。", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard announcement phrasing matches context and raw ASR perfectly." } ], "language": "ja", "packet_ids": [ 81 ], "index": 193, "source_text": "携帯電話での通話は周りのお客様のご迷惑になりますので、お控えください。", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard announcement phrasing matches context and raw ASR perfectly." }, { "start": 1196.4, "end": 1199.5600000000002, "text": "果物、野菜、植物、肉類は、", "raw_asr_text": "果物、野菜、植物、肉類は、", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 81, "words": [ { "word": "果", "start": 1196.4, "end": 1196.76, "probability": 0.9521484375 }, { "word": "物、", "start": 1196.76, "end": 1197.1000000000001, "probability": 0.99951171875 }, { "word": "野", "start": 1197.3600000000001, "end": 1197.6200000000001, "probability": 1.0 }, { "word": "菜、", "start": 1197.6200000000001, "end": 1197.88, "probability": 0.99951171875 }, { "word": "植", "start": 1198.0800000000002, "end": 1198.3200000000002, "probability": 1.0 }, { "word": "物、", "start": 1198.3200000000002, "end": 1198.64, "probability": 1.0 }, { "word": "肉", "start": 1198.88, "end": 1199.1200000000001, "probability": 0.9990234375 }, { "word": "類", "start": 1199.1200000000001, "end": 1199.3000000000002, "probability": 1.0 }, { "word": "は、", "start": 1199.3000000000002, "end": 1199.5600000000002, "probability": 1.0 } ], "asr_metrics": { "avg_logprob": -0.1204914042824193, "compression_ratio": 1.3938053097345133, "no_speech_prob": 5.343340787195006e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "この先、通信用の電波を発する電子機器もご使用になれますが、携帯電話での通話は周りのお客様のご迷惑になりますので、お控えください。果物、野菜、植物、肉類は、税関検査の前に、検疫同植物カウンターでの検査が必要です。" } ], "id": 194, "review_annotation": "", "row_id": 194, "bounded_row_id": 194, "parent_row_ids": [ 194 ], "parent_row_id": 194, "packet_id": 81, "LID_evidence": [ { "region_id": 178, "probabilities": { "ja": 0.9869768252739539, "zh": 0.0008896212412554843, "en": 0.005164397331939723 }, "status": "STABLE" } ], "VAD_region_ids": [ 178 ], "language_confidence": 0.9869768252739539, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1183.9, "packet_end": 1205.5, "raw_text": "果物、野菜、植物、肉類は、", "selected_text": "果物、野菜、植物、肉類は、", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 194 ], "packet_id": 81, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "果物、野菜、植物、肉類は、", "after": "果物、野菜、植物、肉類は、", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard announcement phrasing matches context and raw ASR perfectly." } ], "language": "ja", "packet_ids": [ 81 ], "index": 194, "source_text": "果物、野菜、植物、肉類は、", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard announcement phrasing matches context and raw ASR perfectly." }, { "start": 1200.02, "end": 1204.44, "text": "税関検査の前に、検疫同植物カウンターでの検査が必要です。", "raw_asr_text": "税関検査の前に、検疫同植物カウンターでの検査が必要です。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 81, "words": [ { "word": "税", "start": 1200.02, "end": 1200.18, "probability": 0.9765625 }, { "word": "関", "start": 1200.18, "end": 1200.38, "probability": 0.970703125 }, { "word": "検", "start": 1200.38, "end": 1200.6000000000001, "probability": 1.0 }, { "word": "査", "start": 1200.6000000000001, "end": 1200.74, "probability": 1.0 }, { "word": "の", "start": 1200.74, "end": 1200.8600000000001, "probability": 1.0 }, { "word": "前", "start": 1200.8600000000001, "end": 1201.0800000000002, "probability": 1.0 }, { "word": "に、", "start": 1201.0800000000002, "end": 1201.3600000000001, "probability": 1.0 }, { "word": "検", "start": 1201.48, "end": 1201.64, "probability": 0.91064453125 }, { "word": "疫", "start": 1201.64, "end": 1201.88, "probability": 0.9970703125 }, { "word": "同", "start": 1201.88, "end": 1202.14, "probability": 0.412841796875 }, { "word": "植", "start": 1202.14, "end": 1202.38, "probability": 0.99853515625 }, { "word": "物", "start": 1202.38, "end": 1202.5800000000002, "probability": 1.0 }, { "word": "カ", "start": 1202.5800000000002, "end": 1202.8000000000002, "probability": 1.0 }, { "word": "ウ", "start": 1202.8000000000002, "end": 1202.8600000000001, "probability": 0.99951171875 }, { "word": "ン", "start": 1202.8600000000001, "end": 1202.9, "probability": 0.9970703125 }, { "word": "ター", "start": 1202.9, "end": 1203.02, "probability": 1.0 }, { "word": "で", "start": 1203.02, "end": 1203.18, "probability": 1.0 }, { "word": "の", "start": 1203.18, "end": 1203.3200000000002, "probability": 0.99951171875 }, { "word": "検", "start": 1203.3200000000002, "end": 1203.5800000000002, "probability": 1.0 }, { "word": "査", "start": 1203.5800000000002, "end": 1203.7, "probability": 1.0 }, { "word": "が", "start": 1203.7, "end": 1203.8200000000002, "probability": 1.0 }, { "word": "必", "start": 1203.8200000000002, "end": 1204.02, "probability": 1.0 }, { "word": "要", "start": 1204.02, "end": 1204.18, "probability": 1.0 }, { "word": "です。", "start": 1204.18, "end": 1204.44, "probability": 1.0 } ], "asr_metrics": { "avg_logprob": -0.1204914042824193, "compression_ratio": 1.3938053097345133, "no_speech_prob": 5.343340787195006e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "この先、通信用の電波を発する電子機器もご使用になれますが、携帯電話での通話は周りのお客様のご迷惑になりますので、お控えください。果物、野菜、植物、肉類は、税関検査の前に、検疫同植物カウンターでの検査が必要です。" } ], "id": 195, "review_annotation": "", "row_id": 195, "bounded_row_id": 195, "parent_row_ids": [ 195 ], "parent_row_id": 195, "packet_id": 81, "LID_evidence": [ { "region_id": 178, "probabilities": { "ja": 0.9869768252739539, "zh": 0.0008896212412554843, "en": 0.005164397331939723 }, "status": "STABLE" } ], "VAD_region_ids": [ 178 ], "language_confidence": 0.9869768252739539, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1183.9, "packet_end": 1205.5, "raw_text": "税関検査の前に、検疫同植物カウンターでの検査が必要です。", "selected_text": "税関検査の前に、検疫同植物カウンターでの検査が必要です。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 195 ], "packet_id": 81, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "税関検査の前に、検疫同植物カウンターでの検査が必要です。", "after": "税関検査の前に、検疫同植物カウンターでの検査が必要です。", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard announcement phrasing matches context and raw ASR perfectly." } ], "language": "ja", "packet_ids": [ 81 ], "index": 195, "source_text": "税関検査の前に、検疫同植物カウンターでの検査が必要です。", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard announcement phrasing matches context and raw ASR perfectly." }, { "start": 1208.3, "end": 1212.42, "text": "海外で牛や豚、鳥などの家畜にいる場所にいたお客様は", "raw_asr_text": "海外で牛や豚、鳥などの家畜にいる場所にいたお客様は", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 82, "words": [], "asr_metrics": { "avg_logprob": -0.1438590498531566, "compression_ratio": 1.0887096774193548, "no_speech_prob": 6.546153376307817e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "海外で牛や豚、鳥などの家畜にいる場所にいたお客様は圏域動植物カウンターにお立ち寄りください" } ], "id": 196, "review_annotation": "", "row_id": 196, "bounded_row_id": 196, "parent_row_ids": [ 196 ], "parent_row_id": 196, "packet_id": 82, "LID_evidence": [ { "region_id": 179, "probabilities": { "ja": 0.8881940626435809, "zh": 0.0007941997270487869, "en": 0.003047662075712449 }, "status": "STABLE" } ], "VAD_region_ids": [ 179 ], "language_confidence": 0.8881940626435809, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1207.5, "packet_end": 1217.8, "raw_text": "海外で牛や豚、鳥などの家畜にいる場所にいたお客様は", "selected_text": "海外で牛や豚、鳥などの家畜にいる場所にいたお客様は", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 196 ], "packet_id": 82, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "海外で牛や豚、鳥などの家畜にいる場所にいたお客様は", "after": "海外で牛や豚、鳥などの家畜にいる場所にいたお客様は", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard announcement phrasing matches context and raw ASR perfectly." } ], "language": "ja", "packet_ids": [ 82 ], "index": 196, "source_text": "海外で牛や豚、鳥などの家畜にいる場所にいたお客様は", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard announcement phrasing matches context and raw ASR perfectly.", "raw_word_timestamps": [ { "word": "海", "start": 1208.16, "end": 1208.52, "probability": 0.982421875 }, { "word": "外", "start": 1208.52, "end": 1208.74, "probability": 1.0 }, { "word": "で", "start": 1208.74, "end": 1208.96, "probability": 1.0 }, { "word": "牛", "start": 1208.96, "end": 1209.22, "probability": 0.88427734375 }, { "word": "や", "start": 1209.22, "end": 1209.42, "probability": 0.9990234375 }, { "word": "豚、", "start": 1209.42, "end": 1209.7, "probability": 0.994140625 }, { "word": "鳥", "start": 1209.8, "end": 1210.02, "probability": 0.99462890625 }, { "word": "な", "start": 1210.02, "end": 1210.12, "probability": 0.99951171875 }, { "word": "ど", "start": 1210.12, "end": 1210.26, "probability": 1.0 }, { "word": "の", "start": 1210.26, "end": 1210.36, "probability": 1.0 }, { "word": "家", "start": 1210.36, "end": 1210.52, "probability": 0.9599609375 }, { "word": "畜", "start": 1210.52, "end": 1210.68, "probability": 0.9990234375 }, { "word": "に", "start": 1210.68, "end": 1210.86, "probability": 0.9990234375 }, { "word": "いる", "start": 1210.86, "end": 1210.98, "probability": 0.9990234375 }, { "word": "場", "start": 1210.98, "end": 1211.12, "probability": 0.99951171875 }, { "word": "所", "start": 1211.12, "end": 1211.26, "probability": 1.0 }, { "word": "に", "start": 1211.26, "end": 1211.38, "probability": 1.0 }, { "word": "いた", "start": 1211.38, "end": 1211.58, "probability": 0.919921875 }, { "word": "お", "start": 1211.58, "end": 1211.7, "probability": 0.99951171875 }, { "word": "客", "start": 1211.7, "end": 1211.84, "probability": 1.0 }, { "word": "様", "start": 1211.84, "end": 1212.12, "probability": 0.98974609375 }, { "word": "は", "start": 1212.12, "end": 1212.42, "probability": 1.0 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 1213.38, "end": 1216.44, "text": "圏域動植物カウンターにお立ち寄りください", "raw_asr_text": "圏域動植物カウンターにお立ち寄りください", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 82, "words": [ { "word": "圏", "start": 1213.38, "end": 1213.74, "probability": 0.33642578125 }, { "word": "域", "start": 1213.74, "end": 1214.06, "probability": 0.94140625 }, { "word": "動", "start": 1214.06, "end": 1214.36, "probability": 0.70166015625 }, { "word": "植", "start": 1214.36, "end": 1214.6, "probability": 0.9990234375 }, { "word": "物", "start": 1214.6, "end": 1214.82, "probability": 1.0 }, { "word": "カ", "start": 1214.82, "end": 1215.02, "probability": 1.0 }, { "word": "ウ", "start": 1215.02, "end": 1215.08, "probability": 0.99951171875 }, { "word": "ン", "start": 1215.08, "end": 1215.14, "probability": 0.998046875 }, { "word": "ター", "start": 1215.14, "end": 1215.26, "probability": 1.0 }, { "word": "に", "start": 1215.26, "end": 1215.46, "probability": 1.0 }, { "word": "お", "start": 1215.46, "end": 1215.64, "probability": 0.998046875 }, { "word": "立", "start": 1215.64, "end": 1215.76, "probability": 0.99951171875 }, { "word": "ち", "start": 1215.76, "end": 1215.86, "probability": 0.9990234375 }, { "word": "寄", "start": 1215.86, "end": 1215.96, "probability": 1.0 }, { "word": "り", "start": 1215.96, "end": 1216.06, "probability": 1.0 }, { "word": "ください", "start": 1216.06, "end": 1216.44, "probability": 0.998046875 } ], "asr_metrics": { "avg_logprob": -0.1438590498531566, "compression_ratio": 1.0887096774193548, "no_speech_prob": 6.546153376307817e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "海外で牛や豚、鳥などの家畜にいる場所にいたお客様は圏域動植物カウンターにお立ち寄りください" } ], "id": 197, "review_annotation": "", "row_id": 197, "bounded_row_id": 197, "parent_row_ids": [ 197 ], "parent_row_id": 197, "packet_id": 82, "LID_evidence": [ { "region_id": 180, "probabilities": { "ja": 0.8466494232416153, "zh": 0.0014872718851880304, "en": 0.008560580184816249 }, "status": "STABLE" } ], "VAD_region_ids": [ 180 ], "language_confidence": 0.8466494232416153, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1207.5, "packet_end": 1217.8, "raw_text": "圏域動植物カウンターにお立ち寄りください", "selected_text": "圏域動植物カウンターにお立ち寄りください", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 197 ], "packet_id": 82, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "圏域動植物カウンターにお立ち寄りください", "after": "圏域動植物カウンターにお立ち寄りください", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard announcement phrasing matches context and raw ASR perfectly." } ], "language": "ja", "packet_ids": [ 82 ], "index": 197, "source_text": "圏域動植物カウンターにお立ち寄りください", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard announcement phrasing matches context and raw ASR perfectly." }, { "start": 1224.0, "end": 1228.7, "text": "Ladies and gentlemen, we have landed at Tokyo International Airport.", "raw_asr_text": "Ladies and gentlemen, we have landed at Tokyo International Airport.", "source_language": "en", "source_lang": "en", "asr_packet_id": 83, "words": [], "asr_metrics": { "avg_logprob": -0.2312354530606951, "compression_ratio": 1.0961538461538463, "no_speech_prob": 3.113392182951813e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "en", "text": "Ladies and gentlemen, we have landed at Tokyo International Airport. The local time in Tokyo is 10 minutes past 5." }, { "language": "ja", "text": "Ladies and gentlemen, we have landed at Tokyo International Airport.The local time in Tokyo is 10 minutes past 5." } ], "id": 198, "review_annotation": "", "row_id": 198, "bounded_row_id": 198, "parent_row_ids": [ 198 ], "parent_row_id": 198, "packet_id": 83, "LID_evidence": [ { "region_id": 181, "probabilities": { "ja": 0.16936445286264643, "zh": 0.016301025613211097, "en": 0.668854083120823 }, "status": "STABLE" }, { "region_id": 182, "probabilities": { "ja": 0.27776128507684916, "zh": 0.014762035018065944, "en": 0.48474641516804695 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 181, 182 ], "language_confidence": 0.668854083120823, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1223.2, "packet_end": 1235.8, "raw_text": "Ladies and gentlemen, we have landed at Tokyo International Airport.", "selected_text": "Ladies and gentlemen, we have landed at Tokyo International Airport.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 198 ], "packet_id": 83, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "Ladies and gentlemen, we have landed at Tokyo International Airport.", "after": "Ladies and gentlemen, we have landed at Tokyo International Airport.", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard announcement phrasing matches context and raw ASR perfectly." } ], "language": "en", "packet_ids": [ 83 ], "index": 198, "source_text": "Ladies and gentlemen, we have landed at Tokyo International Airport.", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard announcement phrasing matches context and raw ASR perfectly.", "raw_word_timestamps": [ { "word": " Ladies", "start": 1223.2, "end": 1224.38, "probability": 0.88330078125 }, { "word": " and", "start": 1224.38, "end": 1224.66, "probability": 0.98486328125 }, { "word": " gentlemen,", "start": 1224.66, "end": 1224.92, "probability": 0.79541015625 }, { "word": " we", "start": 1225.16, "end": 1225.22, "probability": 0.9794921875 }, { "word": " have", "start": 1225.22, "end": 1225.42, "probability": 0.9814453125 }, { "word": " landed", "start": 1225.42, "end": 1225.8, "probability": 0.8466796875 }, { "word": " at", "start": 1225.8, "end": 1226.22, "probability": 0.97509765625 }, { "word": " Tokyo", "start": 1226.22, "end": 1227.5800000000002, "probability": 0.9814453125 }, { "word": " International", "start": 1227.5800000000002, "end": 1227.94, "probability": 0.9580078125 }, { "word": " Airport.", "start": 1227.94, "end": 1228.7, "probability": 0.984375 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 1229.3600000000001, "end": 1234.54, "text": "The local time in Tokyo is 10 minutes past 5.", "raw_asr_text": "The local time in Tokyo is 10 minutes past 5.", "source_language": "en", "source_lang": "en", "asr_packet_id": 83, "words": [ { "word": " The", "start": 1229.3600000000001, "end": 1229.96, "probability": 0.98193359375 }, { "word": " local", "start": 1229.96, "end": 1230.24, "probability": 0.98876953125 }, { "word": " time", "start": 1230.24, "end": 1230.5800000000002, "probability": 0.9912109375 }, { "word": " in", "start": 1230.5800000000002, "end": 1230.8600000000001, "probability": 0.99169921875 }, { "word": " Tokyo", "start": 1230.8600000000001, "end": 1231.24, "probability": 0.9951171875 }, { "word": " is", "start": 1231.24, "end": 1231.6200000000001, "probability": 0.99462890625 }, { "word": " 10", "start": 1231.6200000000001, "end": 1233.56, "probability": 0.80224609375 }, { "word": " minutes", "start": 1233.56, "end": 1233.8600000000001, "probability": 0.70556640625 }, { "word": " past", "start": 1233.8600000000001, "end": 1234.2, "probability": 0.99462890625 }, { "word": " 5.", "start": 1234.2, "end": 1234.54, "probability": 0.85595703125 } ], "asr_metrics": { "avg_logprob": -0.2312354530606951, "compression_ratio": 1.0961538461538463, "no_speech_prob": 3.113392182951813e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "en", "text": "Ladies and gentlemen, we have landed at Tokyo International Airport. The local time in Tokyo is 10 minutes past 5." }, { "language": "ja", "text": "Ladies and gentlemen, we have landed at Tokyo International Airport.The local time in Tokyo is 10 minutes past 5." } ], "id": 199, "review_annotation": "", "row_id": 199, "bounded_row_id": 199, "parent_row_ids": [ 199 ], "parent_row_id": 199, "packet_id": 83, "LID_evidence": [ { "region_id": 183, "probabilities": { "ja": 0.007072872528806329, "zh": 0.0049729639198631045, "en": 0.8060757279396057 }, "status": "STABLE" }, { "region_id": 184, "probabilities": { "ja": 0, "zh": 0, "en": 0 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 183, 184 ], "language_confidence": 0.8060757279396057, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1223.2, "packet_end": 1235.8, "raw_text": "The local time in Tokyo is 10 minutes past 5.", "selected_text": "The local time in Tokyo is 10 minutes past 5.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 199 ], "packet_id": 83, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "The local time in Tokyo is 10 minutes past 5.", "after": "The local time in Tokyo is 10 minutes past 5.", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard announcement phrasing matches context and raw ASR perfectly." } ], "language": "en", "packet_ids": [ 83 ], "index": 199, "source_text": "The local time in Tokyo is 10 minutes past 5.", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard announcement phrasing matches context and raw ASR perfectly." }, { "start": 1238.6, "end": 1244.44, "text": "March 14th. Please remain seated until the captain turns off the seatbelt sign.", "raw_asr_text": "March 14th. Please image seated until the captain turns off the seatbelt sign.", "source_language": "en", "source_lang": "en", "asr_packet_id": 84, "words": [], "asr_metrics": { "avg_logprob": -0.2525431798852008, "compression_ratio": 1.3108108108108107, "no_speech_prob": 5.4256946158126595e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "en", "text": "March 14th. Please image seated until the captain turns off the seatbelt sign. Take care when opening the overhead compartment. From now on, you can use electronic devices that emit radio waves." }, { "language": "ja", "text": "MAR 14Please, you may see it until the captain turns off the seatbelt signTake care when opening the overhead compartmentFrom now on, you can use electronic devices that emit radio waves" } ], "id": 200, "review_annotation": "", "row_id": 200, "bounded_row_id": 200, "parent_row_ids": [ 200 ], "parent_row_id": 200, "packet_id": 84, "LID_evidence": [ { "region_id": 185, "probabilities": { "ja": 0.0929203995813926, "zh": 0.03486240406831106, "en": 0.57954208056132 }, "status": "UNCERTAIN" }, { "region_id": 186, "probabilities": { "ja": 0.00834626511884077, "zh": 0.011842171913788965, "en": 0.907352680961291 }, "status": "STABLE" } ], "VAD_region_ids": [ 185, 186 ], "language_confidence": 0.907352680961291, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1237.8, "packet_end": 1254.7, "raw_text": "March 14th. Please image seated until the captain turns off the seatbelt sign.", "selected_text": "March 14th. Please remain seated until the captain turns off the seatbelt sign.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 200 ], "packet_id": 84, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "March 14th. Please image seated until the captain turns off the seatbelt sign.", "after": "March 14th. Please remain seated until the captain turns off the seatbelt sign.", "classification": "CERTAIN_RESTORATION", "reason": "Source 'image' is an ASR error for 'remain', confirmed by independent ASR." } ], "language": "en", "packet_ids": [ 84 ], "index": 200, "independent_ASR": { "text": "March 14th. Please remain seated until the captain turns off the seatbelt sign. Take care when opening the overhead compartment. Column one: You can use electronic devices that emit radio waves.", "chunks": [ { "text": "March 14th. Please remain seated until the captain turns off the seatbelt sign. Take care when opening the overhead compartment. Column one: You can use electronic devices that emit radio waves.", "start": 0.0, "end": 16.9, "chunk_index": 0, "language": "English", "finish_reason": "eos", "truncated": false, "generated_tokens": 40, "max_new_tokens": 235 } ], "detected_language": "English", "segments": [ { "start": 1237.8, "end": 1254.7, "text": "March 14th. Please remain seated until the captain turns off the seatbelt sign. Take care when opening the overhead compartment. Column one: You can use electronic devices that emit radio waves." } ], "start": 1237.8, "end": 1254.7, "engine": "mlx-qwen3-asr", "language": "en" }, "raw_word_timestamps": [ { "word": " March", "start": 1237.8, "end": 1238.9199999999998, "probability": 0.76171875 }, { "word": " 14th.", "start": 1238.9199999999998, "end": 1239.8999999999999, "probability": 0.7490234375 }, { "word": " Please", "start": 1240.6, "end": 1241.32, "probability": 0.9873046875 }, { "word": " image", "start": 1241.32, "end": 1241.68, "probability": 0.1595458984375 }, { "word": " seated", "start": 1241.68, "end": 1242.04, "probability": 0.98193359375 }, { "word": " until", "start": 1242.04, "end": 1242.3999999999999, "probability": 0.98046875 }, { "word": " the", "start": 1242.3999999999999, "end": 1242.6, "probability": 0.994140625 }, { "word": " captain", "start": 1242.6, "end": 1242.8799999999999, "probability": 0.81689453125 }, { "word": " turns", "start": 1242.8799999999999, "end": 1243.3, "probability": 0.9970703125 }, { "word": " off", "start": 1243.3, "end": 1243.48, "probability": 0.95068359375 }, { "word": " the", "start": 1243.48, "end": 1243.62, "probability": 0.9951171875 }, { "word": " seatbelt", "start": 1243.62, "end": 1244.04, "probability": 0.87548828125 }, { "word": " sign.", "start": 1244.04, "end": 1244.44, "probability": 0.998046875 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "March 14th. Please remain seated until the captain turns off the seatbelt sign.", "review_status": "CERTAIN_RESTORATION", "classification": "CERTAIN_RESTORATION", "confidence": 0.95, "correction_reason": "Source 'image' is an ASR error for 'remain', confirmed by independent ASR." }, { "start": 1244.96, "end": 1247.84, "text": "Take care when opening the overhead compartment.", "raw_asr_text": "Take care when opening the overhead compartment.", "source_language": "en", "source_lang": "en", "asr_packet_id": 84, "words": [ { "word": " Take", "start": 1244.96, "end": 1245.68, "probability": 0.99560546875 }, { "word": " care", "start": 1245.68, "end": 1245.94, "probability": 0.9990234375 }, { "word": " when", "start": 1245.94, "end": 1246.1599999999999, "probability": 0.9443359375 }, { "word": " opening", "start": 1246.1599999999999, "end": 1246.5, "probability": 0.99755859375 }, { "word": " the", "start": 1246.5, "end": 1246.74, "probability": 0.99853515625 }, { "word": " overhead", "start": 1246.74, "end": 1247.18, "probability": 0.98828125 }, { "word": " compartment.", "start": 1247.18, "end": 1247.84, "probability": 0.9765625 } ], "asr_metrics": { "avg_logprob": -0.2525431798852008, "compression_ratio": 1.3108108108108107, "no_speech_prob": 5.4256946158126595e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "en", "text": "March 14th. Please image seated until the captain turns off the seatbelt sign. Take care when opening the overhead compartment. From now on, you can use electronic devices that emit radio waves." }, { "language": "ja", "text": "MAR 14Please, you may see it until the captain turns off the seatbelt signTake care when opening the overhead compartmentFrom now on, you can use electronic devices that emit radio waves" } ], "id": 201, "review_annotation": "", "row_id": 201, "bounded_row_id": 201, "parent_row_ids": [ 201 ], "parent_row_id": 201, "packet_id": 84, "LID_evidence": [ { "region_id": 187, "probabilities": { "ja": 0.006771461333846674, "zh": 0.011271947878412902, "en": 0.8795683532953262 }, "status": "STABLE" } ], "VAD_region_ids": [ 187 ], "language_confidence": 0.8795683532953262, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1237.8, "packet_end": 1254.7, "raw_text": "Take care when opening the overhead compartment.", "selected_text": "Take care when opening the overhead compartment.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 201 ], "packet_id": 84, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "Take care when opening the overhead compartment.", "after": "Take care when opening the overhead compartment.", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Matches standard safety announcement wording and context perfectly." } ], "language": "en", "packet_ids": [ 84 ], "index": 201, "source_text": "Take care when opening the overhead compartment.", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Matches standard safety announcement wording and context perfectly." }, { "start": 1248.82, "end": 1253.58, "text": "From now on, you can use electronic devices that emit radio waves.", "raw_asr_text": "From now on, you can use electronic devices that emit radio waves.", "source_language": "en", "source_lang": "en", "asr_packet_id": 84, "words": [ { "word": " From", "start": 1248.82, "end": 1249.54, "probability": 0.9248046875 }, { "word": " now", "start": 1249.54, "end": 1249.7, "probability": 0.88037109375 }, { "word": " on,", "start": 1249.7, "end": 1249.94, "probability": 0.9921875 }, { "word": " you", "start": 1250.04, "end": 1250.1399999999999, "probability": 0.99609375 }, { "word": " can", "start": 1250.1399999999999, "end": 1250.34, "probability": 0.99951171875 }, { "word": " use", "start": 1250.34, "end": 1250.6599999999999, "probability": 0.994140625 }, { "word": " electronic", "start": 1250.6599999999999, "end": 1251.06, "probability": 0.990234375 }, { "word": " devices", "start": 1251.06, "end": 1251.6599999999999, "probability": 0.99853515625 }, { "word": " that", "start": 1251.6599999999999, "end": 1252.3799999999999, "probability": 0.98779296875 }, { "word": " emit", "start": 1252.3799999999999, "end": 1252.6599999999999, "probability": 0.97265625 }, { "word": " radio", "start": 1252.6599999999999, "end": 1253.12, "probability": 0.97900390625 }, { "word": " waves.", "start": 1253.12, "end": 1253.58, "probability": 0.99951171875 } ], "asr_metrics": { "avg_logprob": -0.2525431798852008, "compression_ratio": 1.3108108108108107, "no_speech_prob": 5.4256946158126595e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "en", "text": "March 14th. Please image seated until the captain turns off the seatbelt sign. Take care when opening the overhead compartment. From now on, you can use electronic devices that emit radio waves." }, { "language": "ja", "text": "MAR 14Please, you may see it until the captain turns off the seatbelt signTake care when opening the overhead compartmentFrom now on, you can use electronic devices that emit radio waves" } ], "id": 202, "review_annotation": "", "row_id": 202, "bounded_row_id": 202, "parent_row_ids": [ 202 ], "parent_row_id": 202, "packet_id": 84, "LID_evidence": [ { "region_id": 188, "probabilities": { "ja": 0.15510787339881063, "zh": 0.032431098795495925, "en": 0.534100541472435 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 188 ], "language_confidence": 0.534100541472435, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1237.8, "packet_end": 1254.7, "raw_text": "From now on, you can use electronic devices that emit radio waves.", "selected_text": "From now on, you can use electronic devices that emit radio waves.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 202 ], "packet_id": 84, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "From now on, you can use electronic devices that emit radio waves.", "after": "From now on, you can use electronic devices that emit radio waves.", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Matches standard post-landing announcement wording and context." } ], "language": "en", "packet_ids": [ 84 ], "index": 202, "source_text": "From now on, you can use electronic devices that emit radio waves.", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Matches standard post-landing announcement wording and context." }, { "start": 1258.2, "end": 1263.26, "text": "However, please refrain from talking on cell phones out of consideration for other passengers.", "raw_asr_text": "However, please stay fair from talking on cell phones out of consideration for other passengers.", "source_language": "en", "source_lang": "en", "asr_packet_id": 85, "words": [], "asr_metrics": { "avg_logprob": -0.32039632797241213, "compression_ratio": 1.1162790697674418, "no_speech_prob": 4.9406770341597905e-11 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "en", "text": "However, please stay fair from talking on cell phones out of consideration for other passengers." }, { "language": "ja", "text": "However, please stay fair from talking on cell phones out of consideration for other passengers." } ], "id": 203, "review_annotation": "", "row_id": 203, "bounded_row_id": 203, "parent_row_ids": [ 203 ], "parent_row_id": 203, "packet_id": 85, "LID_evidence": [ { "region_id": 189, "probabilities": { "ja": 0.09028495731763542, "zh": 0.0074660079553723335, "en": 0.6142670214176178 }, "status": "STABLE" }, { "region_id": 190, "probabilities": { "ja": 0.10518697245667379, "zh": 0.006420172052457929, "en": 0.6194887260595957 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 189, 190 ], "language_confidence": 0.6194887260595957, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1257.4, "packet_end": 1264.6, "raw_text": "However, please stay fair from talking on cell phones out of consideration for other passengers.", "selected_text": "However, please refrain from talking on cell phones out of consideration for other passengers.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 203 ], "packet_id": 85, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "However, please stay fair from talking on cell phones out of consideration for other passengers.", "after": "However, please refrain from talking on cell phones out of consideration for other passengers.", "classification": "CERTAIN_RESTORATION", "reason": "Source 'stay fair' is an ASR error for 'refrain', confirmed by independent ASR." } ], "language": "en", "packet_ids": [ 85 ], "index": 203, "independent_ASR": { "text": "However, please refrain from talking on cell phones out of consideration for other passengers.", "chunks": [ { "text": "However, please refrain from talking on cell phones out of consideration for other passengers.", "start": 0.0, "end": 7.2, "chunk_index": 0, "language": "English", "finish_reason": "eos", "truncated": false, "generated_tokens": 16, "max_new_tokens": 128 } ], "detected_language": "English", "segments": [ { "start": 1257.4, "end": 1264.6000000000001, "text": "However, please refrain from talking on cell phones out of consideration for other passengers." } ], "start": 1257.4, "end": 1264.6, "engine": "mlx-qwen3-asr", "language": "en" }, "raw_word_timestamps": [ { "word": " However,", "start": 1258.02, "end": 1258.5800000000002, "probability": 0.8408203125 }, { "word": " please", "start": 1258.8600000000001, "end": 1259.0, "probability": 0.9921875 }, { "word": " stay", "start": 1259.0, "end": 1259.18, "probability": 0.91943359375 }, { "word": " fair", "start": 1259.18, "end": 1259.4, "probability": 0.64208984375 }, { "word": " from", "start": 1259.4, "end": 1259.6000000000001, "probability": 0.88037109375 }, { "word": " talking", "start": 1259.6000000000001, "end": 1259.88, "probability": 0.97314453125 }, { "word": " on", "start": 1259.88, "end": 1260.14, "probability": 0.97998046875 }, { "word": " cell", "start": 1260.14, "end": 1260.42, "probability": 0.837890625 }, { "word": " phones", "start": 1260.42, "end": 1260.74, "probability": 0.94970703125 }, { "word": " out", "start": 1260.74, "end": 1261.3600000000001, "probability": 0.56689453125 }, { "word": " of", "start": 1261.3600000000001, "end": 1261.52, "probability": 0.9921875 }, { "word": " consideration", "start": 1261.52, "end": 1262.02, "probability": 0.783203125 }, { "word": " for", "start": 1262.02, "end": 1262.44, "probability": 0.97607421875 }, { "word": " other", "start": 1262.44, "end": 1262.68, "probability": 0.986328125 }, { "word": " passengers.", "start": 1262.68, "end": 1263.26, "probability": 0.99658203125 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "However, please refrain from talking on cell phones out of consideration for other passengers.", "review_status": "CERTAIN_RESTORATION", "classification": "CERTAIN_RESTORATION", "confidence": 0.95, "correction_reason": "Source 'stay fair' is an ASR error for 'refrain', confirmed by independent ASR." }, { "start": 1275.3, "end": 1282.08, "text": "Passengers carrying fresh fruit, vegetables, plant and meat products must proceed to the", "raw_asr_text": "Passengers carrying fresh fruit, vegetables, plant and meat products must proceed to the", "source_language": "en", "source_lang": "en", "asr_packet_id": 86, "words": [], "asr_metrics": { "avg_logprob": -0.2796118951612903, "compression_ratio": 1.2956521739130435, "no_speech_prob": 2.839717870573022e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "en", "text": "Passengers carrying fresh fruit, vegetables, plant and meat products must proceed to the plant and animal quarantine counter before going to customs." } ], "id": 204, "review_annotation": "", "row_id": 204, "bounded_row_id": 204, "parent_row_ids": [ 204 ], "parent_row_id": 204, "packet_id": 86, "LID_evidence": [ { "region_id": 191, "probabilities": { "ja": 0.004361105060929226, "zh": 0.0021908572234679013, "en": 0.8979826304647658 }, "status": "STABLE" }, { "region_id": 192, "probabilities": { "ja": 0.008370449909797082, "zh": 0.003698318135530616, "en": 0.8752331000107986 }, "status": "STABLE" } ], "VAD_region_ids": [ 191, 192 ], "language_confidence": 0.8979826304647658, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1274.5, "packet_end": 1286.6, "raw_text": "Passengers carrying fresh fruit, vegetables, plant and meat products must proceed to the", "selected_text": "Passengers carrying fresh fruit, vegetables, plant and meat products must proceed to the", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 204 ], "packet_id": 86, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "Passengers carrying fresh fruit, vegetables, plant and meat products must proceed to the", "after": "Passengers carrying fresh fruit, vegetables, plant and meat products must proceed to the", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Correct partial sentence matching standard quarantine announcement context." } ], "language": "en", "packet_ids": [ 86 ], "index": 204, "source_text": "Passengers carrying fresh fruit, vegetables, plant and meat products must proceed to the", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Correct partial sentence matching standard quarantine announcement context.", "raw_word_timestamps": [ { "word": " Passengers", "start": 1275.28, "end": 1276.04, "probability": 0.96142578125 }, { "word": " carrying", "start": 1276.04, "end": 1276.62, "probability": 0.99658203125 }, { "word": " fresh", "start": 1276.62, "end": 1277.14, "probability": 0.9111328125 }, { "word": " fruit,", "start": 1277.14, "end": 1277.48, "probability": 0.82763671875 }, { "word": " vegetables,", "start": 1277.7, "end": 1278.02, "probability": 0.984375 }, { "word": " plant", "start": 1278.46, "end": 1278.72, "probability": 0.736328125 }, { "word": " and", "start": 1278.72, "end": 1279.08, "probability": 0.76171875 }, { "word": " meat", "start": 1279.08, "end": 1280.08, "probability": 0.9853515625 }, { "word": " products", "start": 1280.08, "end": 1280.66, "probability": 0.9833984375 }, { "word": " must", "start": 1280.66, "end": 1281.34, "probability": 0.96435546875 }, { "word": " proceed", "start": 1281.34, "end": 1281.76, "probability": 0.99560546875 }, { "word": " to", "start": 1281.76, "end": 1281.92, "probability": 0.9951171875 }, { "word": " the", "start": 1281.92, "end": 1282.08, "probability": 0.96923828125 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 1282.08, "end": 1285.26, "text": "plant and animal quarantine counter before going to customs.", "raw_asr_text": "plant and animal quarantine counter before going to customs.", "source_language": "en", "source_lang": "en", "asr_packet_id": 86, "words": [ { "word": " plant", "start": 1282.08, "end": 1282.32, "probability": 0.90087890625 }, { "word": " and", "start": 1282.32, "end": 1282.54, "probability": 0.9189453125 }, { "word": " animal", "start": 1282.54, "end": 1282.92, "probability": 0.9921875 }, { "word": " quarantine", "start": 1282.92, "end": 1283.46, "probability": 0.99658203125 }, { "word": " counter", "start": 1283.46, "end": 1283.92, "probability": 0.9775390625 }, { "word": " before", "start": 1283.92, "end": 1284.24, "probability": 0.99072265625 }, { "word": " going", "start": 1284.24, "end": 1284.56, "probability": 0.9951171875 }, { "word": " to", "start": 1284.56, "end": 1284.8, "probability": 0.99609375 }, { "word": " customs.", "start": 1284.8, "end": 1285.26, "probability": 0.978515625 } ], "asr_metrics": { "avg_logprob": -0.2796118951612903, "compression_ratio": 1.2956521739130435, "no_speech_prob": 2.839717870573022e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "en", "text": "Passengers carrying fresh fruit, vegetables, plant and meat products must proceed to the plant and animal quarantine counter before going to customs." } ], "id": 205, "review_annotation": "", "row_id": 205, "bounded_row_id": 205, "parent_row_ids": [ 205 ], "parent_row_id": 205, "packet_id": 86, "LID_evidence": [ { "region_id": 192, "probabilities": { "ja": 0.008370449909797082, "zh": 0.003698318135530616, "en": 0.8752331000107986 }, "status": "STABLE" } ], "VAD_region_ids": [ 192 ], "language_confidence": 0.8752331000107986, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1274.5, "packet_end": 1286.6, "raw_text": "plant and animal quarantine counter before going to customs.", "selected_text": "plant and animal quarantine counter before going to customs.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 205 ], "packet_id": 86, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "plant and animal quarantine counter before going to customs.", "after": "plant and animal quarantine counter before going to customs.", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Correct completion of the previous sentence matching standard context." } ], "language": "en", "packet_ids": [ 86 ], "index": 205, "source_text": "plant and animal quarantine counter before going to customs.", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Correct completion of the previous sentence matching standard context." }, { "start": 1288.9, "end": 1301.2400000000002, "text": "If you have visited any overseas livestock facilities or may come into contact with livestock in Japan, please proceed to the plant and animal quarantine counter.", "raw_asr_text": "If you have visited any overseas livestock facilities or may come into contact with livestock in Japan, please proceed to the plant and animal quarantine counter.", "source_language": "en", "source_lang": "en", "asr_packet_id": 87, "words": [], "asr_metrics": { "avg_logprob": -0.21895979296776555, "compression_ratio": 1.3388429752066116, "no_speech_prob": 5.102852515537215e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "en", "text": "If you have visited any overseas livestock facilities or may come into contact with livestock in Japan, please proceed to the plant and animal quarantine counter." } ], "id": 206, "review_annotation": "", "row_id": 206, "bounded_row_id": 206, "parent_row_ids": [ 206 ], "parent_row_id": 206, "packet_id": 87, "LID_evidence": [ { "region_id": 193, "probabilities": { "ja": 0.006818602187559008, "zh": 0.006009061960503459, "en": 0.6893138311803341 }, "status": "STABLE" }, { "region_id": 194, "probabilities": { "ja": 0.04527103365928104, "zh": 0.01507123209409077, "en": 0.8782111297954213 }, "status": "STABLE" } ], "VAD_region_ids": [ 193, 194 ], "language_confidence": 0.8782111297954213, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1288.1000000000001, "packet_end": 1302.5, "raw_text": "If you have visited any overseas livestock facilities or may come into contact with livestock in Japan, please proceed to the plant and animal quarantine counter.", "selected_text": "If you have visited any overseas livestock facilities or may come into contact with livestock in Japan, please proceed to the plant and animal quarantine counter.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 206 ], "packet_id": 87, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "If you have visited any overseas livestock facilities or may come into contact with livestock in Japan, please proceed to the plant and animal quarantine counter.", "after": "If you have visited any overseas livestock facilities or may come into contact with livestock in Japan, please proceed to the plant and animal quarantine counter.", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Matches standard quarantine announcement wording and context perfectly." } ], "language": "en", "packet_ids": [ 87 ], "index": 206, "source_text": "If you have visited any overseas livestock facilities or may come into contact with livestock in Japan, please proceed to the plant and animal quarantine counter.", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Matches standard quarantine announcement wording and context perfectly.", "raw_word_timestamps": [ { "word": " If", "start": 1288.1000000000001, "end": 1289.2, "probability": 0.437255859375 }, { "word": " you", "start": 1289.2, "end": 1289.2600000000002, "probability": 0.9951171875 }, { "word": " have", "start": 1289.2600000000002, "end": 1289.4800000000002, "probability": 0.974609375 }, { "word": " visited", "start": 1289.4800000000002, "end": 1289.88, "probability": 0.9853515625 }, { "word": " any", "start": 1289.88, "end": 1290.3200000000002, "probability": 0.87109375 }, { "word": " overseas", "start": 1290.3200000000002, "end": 1291.0600000000002, "probability": 0.96875 }, { "word": " livestock", "start": 1291.0600000000002, "end": 1291.5400000000002, "probability": 0.95849609375 }, { "word": " facilities", "start": 1291.5400000000002, "end": 1292.3600000000001, "probability": 0.96728515625 }, { "word": " or", "start": 1292.3600000000001, "end": 1294.8400000000001, "probability": 0.59423828125 }, { "word": " may", "start": 1294.8400000000001, "end": 1295.0600000000002, "probability": 0.9736328125 }, { "word": " come", "start": 1295.0600000000002, "end": 1295.2800000000002, "probability": 0.97705078125 }, { "word": " into", "start": 1295.2800000000002, "end": 1295.5400000000002, "probability": 0.85498046875 }, { "word": " contact", "start": 1295.5400000000002, "end": 1295.9800000000002, "probability": 0.99755859375 }, { "word": " with", "start": 1295.9800000000002, "end": 1296.22, "probability": 0.99462890625 }, { "word": " livestock", "start": 1296.22, "end": 1296.6200000000001, "probability": 0.9833984375 }, { "word": " in", "start": 1296.6200000000001, "end": 1296.96, "probability": 0.98974609375 }, { "word": " Japan,", "start": 1296.96, "end": 1297.3000000000002, "probability": 0.98388671875 }, { "word": " please", "start": 1297.6200000000001, "end": 1298.0600000000002, "probability": 0.9677734375 }, { "word": " proceed", "start": 1298.0600000000002, "end": 1298.72, "probability": 0.99169921875 }, { "word": " to", "start": 1298.72, "end": 1298.88, "probability": 0.99267578125 }, { "word": " the", "start": 1298.88, "end": 1298.9800000000002, "probability": 0.91943359375 }, { "word": " plant", "start": 1298.9800000000002, "end": 1299.2400000000002, "probability": 0.76123046875 }, { "word": " and", "start": 1299.2400000000002, "end": 1299.4800000000002, "probability": 0.8857421875 }, { "word": " animal", "start": 1299.4800000000002, "end": 1299.88, "probability": 0.984375 }, { "word": " quarantine", "start": 1299.88, "end": 1300.68, "probability": 0.99072265625 }, { "word": " counter.", "start": 1300.68, "end": 1301.2400000000002, "probability": 0.90380859375 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 1305.1, "end": 1310.34, "text": "Thank you for flying with ANA and Star Alliance partner. We look forward to serving you again.", "raw_asr_text": "Thank you for flying with ANA and Science Partner. We look forward to serving you again.", "source_language": "en", "source_lang": "en", "asr_packet_id": 88, "words": [], "asr_metrics": { "avg_logprob": -0.2998306317762895, "compression_ratio": 1.0476190476190477, "no_speech_prob": 4.429774255743091e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "en", "text": "Thank you for flying with ANA and Science Partner. We look forward to serving you again." } ], "id": 207, "review_annotation": "", "row_id": 207, "bounded_row_id": 207, "parent_row_ids": [ 207 ], "parent_row_id": 207, "packet_id": 88, "LID_evidence": [ { "region_id": 195, "probabilities": { "ja": 0.0028420991979406346, "zh": 0.0027770669611137627, "en": 0.8420769359384265 }, "status": "STABLE" }, { "region_id": 196, "probabilities": { "ja": 0.005845617027565216, "zh": 0.0020662828246713616, "en": 0.9670389493306478 }, "status": "STABLE" } ], "VAD_region_ids": [ 195, 196 ], "language_confidence": 0.9670389493306478, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1304.3, "packet_end": 1311.7, "raw_text": "Thank you for flying with ANA and Science Partner. We look forward to serving you again.", "selected_text": "Thank you for flying with ANA and Star Alliance partner. We look forward to serving you again.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 207 ], "packet_id": 88, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "Thank you for flying with ANA and Science Partner. We look forward to serving you again.", "after": "Thank you for flying with ANA and Star Alliance partner. We look forward to serving you again.", "classification": "CERTAIN_RESTORATION", "reason": "Independent ASR and Japanese context confirm 'Star Alliance' over 'Science Partner'." } ], "language": "en", "packet_ids": [ 88 ], "index": 207, "independent_ASR": { "text": "Thank you for flying with ANA and Star Alliance partner. We look forward to serving you again.", "chunks": [ { "text": "Thank you for flying with ANA and Star Alliance partner. We look forward to serving you again.", "start": 0.0, "end": 7.4, "chunk_index": 0, "language": "English", "finish_reason": "eos", "truncated": false, "generated_tokens": 20, "max_new_tokens": 128 } ], "detected_language": "English", "segments": [ { "start": 1304.3, "end": 1311.7, "text": "Thank you for flying with ANA and Star Alliance partner. We look forward to serving you again." } ], "start": 1304.3, "end": 1311.7, "engine": "mlx-qwen3-asr", "language": "en" }, "raw_word_timestamps": [ { "word": " Thank", "start": 1304.3, "end": 1305.34, "probability": 0.84033203125 }, { "word": " you", "start": 1305.34, "end": 1305.54, "probability": 0.97802734375 }, { "word": " for", "start": 1305.54, "end": 1305.74, "probability": 0.97509765625 }, { "word": " flying", "start": 1305.74, "end": 1306.02, "probability": 0.98828125 }, { "word": " with", "start": 1306.02, "end": 1306.26, "probability": 0.97607421875 }, { "word": " ANA", "start": 1306.26, "end": 1306.84, "probability": 0.75244140625 }, { "word": " and", "start": 1306.84, "end": 1307.12, "probability": 0.91064453125 }, { "word": " Science", "start": 1307.12, "end": 1307.3799999999999, "probability": 0.54931640625 }, { "word": " Partner.", "start": 1307.3799999999999, "end": 1307.8999999999999, "probability": 0.529296875 }, { "word": " We", "start": 1308.28, "end": 1308.7, "probability": 0.98193359375 }, { "word": " look", "start": 1308.7, "end": 1308.8799999999999, "probability": 0.9853515625 }, { "word": " forward", "start": 1308.8799999999999, "end": 1309.22, "probability": 0.99609375 }, { "word": " to", "start": 1309.22, "end": 1309.44, "probability": 0.99462890625 }, { "word": " serving", "start": 1309.44, "end": 1309.8, "probability": 0.818359375 }, { "word": " you", "start": 1309.8, "end": 1310.0, "probability": 0.97998046875 }, { "word": " again.", "start": 1310.0, "end": 1310.34, "probability": 0.99609375 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "Thank you for flying with ANA and Star Alliance partner. We look forward to serving you again.", "review_status": "CERTAIN_RESTORATION", "classification": "CERTAIN_RESTORATION", "confidence": 0.98, "correction_reason": "Independent ASR and Japanese context confirm 'Star Alliance' over 'Science Partner'." }, { "start": 1314.92, "end": 1320.38, "text": "皆さま、今日もANAとスターライアンスパートナーとのコードシェア便をご利用いただきましてありがとうございました。", "raw_asr_text": "皆さま、今日もANAとスターライアンスパートナーとのコードシェア便をご利用いただきましてありがとうございました。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 89, "words": [ { "word": "皆", "start": 1314.92, "end": 1315.2, "probability": 0.9892578125 }, { "word": "さ", "start": 1315.2, "end": 1315.3200000000002, "probability": 0.64794921875 }, { "word": "ま、", "start": 1315.3200000000002, "end": 1315.46, "probability": 0.99951171875 }, { "word": "今日", "start": 1315.6000000000001, "end": 1315.7, "probability": 0.95654296875 }, { "word": "も", "start": 1315.7, "end": 1315.94, "probability": 1.0 }, { "word": "ANA", "start": 1315.94, "end": 1316.2600000000002, "probability": 0.47216796875 }, { "word": "と", "start": 1316.2600000000002, "end": 1316.5200000000002, "probability": 0.99169921875 }, { "word": "ス", "start": 1316.5200000000002, "end": 1316.7, "probability": 0.95166015625 }, { "word": "ター", "start": 1316.7, "end": 1316.7800000000002, "probability": 0.998046875 }, { "word": "ライ", "start": 1316.7800000000002, "end": 1316.96, "probability": 0.98974609375 }, { "word": "ア", "start": 1316.96, "end": 1317.0800000000002, "probability": 0.97216796875 }, { "word": "ン", "start": 1317.0800000000002, "end": 1317.16, "probability": 0.93359375 }, { "word": "ス", "start": 1317.16, "end": 1317.18, "probability": 0.99267578125 }, { "word": "パ", "start": 1317.18, "end": 1317.3200000000002, "probability": 0.9931640625 }, { "word": "ート", "start": 1317.3200000000002, "end": 1317.4800000000002, "probability": 0.99951171875 }, { "word": "ナ", "start": 1317.4800000000002, "end": 1317.6000000000001, "probability": 0.99951171875 }, { "word": "ー", "start": 1317.6000000000001, "end": 1317.68, "probability": 0.99951171875 }, { "word": "と", "start": 1317.68, "end": 1317.7800000000002, "probability": 0.9990234375 }, { "word": "の", "start": 1317.7800000000002, "end": 1317.9, "probability": 1.0 }, { "word": "コ", "start": 1317.9, "end": 1318.0400000000002, "probability": 0.7001953125 }, { "word": "ード", "start": 1318.0400000000002, "end": 1318.2, "probability": 0.99853515625 }, { "word": "シ", "start": 1318.2, "end": 1318.3400000000001, "probability": 0.81640625 }, { "word": "ェ", "start": 1318.3400000000001, "end": 1318.42, "probability": 0.91845703125 }, { "word": "ア", "start": 1318.42, "end": 1318.4800000000002, "probability": 0.99951171875 }, { "word": "便", "start": 1318.4800000000002, "end": 1318.5800000000002, "probability": 0.87646484375 }, { "word": "を", "start": 1318.5800000000002, "end": 1318.8000000000002, "probability": 0.99951171875 }, { "word": "ご", "start": 1318.8000000000002, "end": 1318.92, "probability": 0.98974609375 }, { "word": "利", "start": 1318.92, "end": 1319.0400000000002, "probability": 1.0 }, { "word": "用", "start": 1319.0400000000002, "end": 1319.16, "probability": 1.0 }, { "word": "いただ", "start": 1319.16, "end": 1319.3200000000002, "probability": 0.99951171875 }, { "word": "き", "start": 1319.3200000000002, "end": 1319.5400000000002, "probability": 1.0 }, { "word": "ま", "start": 1319.5400000000002, "end": 1319.66, "probability": 0.99853515625 }, { "word": "して", "start": 1319.66, "end": 1319.92, "probability": 1.0 }, { "word": "ありがとうございました。", "start": 1319.92, "end": 1320.38, "probability": 0.78955078125 } ], "asr_metrics": { "avg_logprob": -0.1757991008269481, "compression_ratio": 1.2366412213740459, "no_speech_prob": 5.363948868408741e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "皆さま、今日もANAとスターライアンスパートナーとのコードシェア便をご利用いただきましてありがとうございました。" } ], "id": 208, "review_annotation": "", "row_id": 208, "bounded_row_id": 208, "parent_row_ids": [ 208 ], "parent_row_id": 208, "packet_id": 89, "LID_evidence": [ { "region_id": 197, "probabilities": { "ja": 0.9843273222446441, "zh": 0.0009973951993742958, "en": 0.0061510401195846495 }, "status": "STABLE" } ], "VAD_region_ids": [ 197 ], "language_confidence": 0.9843273222446441, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1314.1000000000001, "packet_end": 1322.3, "raw_text": "皆さま、今日もANAとスターライアンスパートナーとのコードシェア便をご利用いただきましてありがとうございました。", "selected_text": "皆さま、今日もANAとスターライアンスパートナーとのコードシェア便をご利用いただきましてありがとうございました。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 208 ], "packet_id": 89, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "皆さま、今日もANAとスターライアンスパートナーとのコードシェア便をご利用いただきましてありがとうございました。", "after": "皆さま、今日もANAとスターライアンスパートナーとのコードシェア便をご利用いただきましてありがとうございました。", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Matches standard closing announcement wording and context." } ], "language": "ja", "packet_ids": [ 89 ], "index": 208, "source_text": "皆さま、今日もANAとスターライアンスパートナーとのコードシェア便をご利用いただきましてありがとうございました。", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Matches standard closing announcement wording and context." }, { "start": 1324.2, "end": 1326.3200000000002, "text": "次のご利用をお待ちしております", "raw_asr_text": "次のご登場をお待ちしております", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 90, "words": [], "asr_metrics": { "avg_logprob": -0.2513828913370768, "compression_ratio": 0.8333333333333334, "no_speech_prob": 5.156168894626667e-12 }, "lid_status": "UNCERTAIN", "alternatives": [ { "language": "ja", "text": "次のご登場をお待ちしております" } ], "id": 209, "review_annotation": "待確認", "row_id": 209, "bounded_row_id": 209, "parent_row_ids": [ 209 ], "parent_row_id": 209, "packet_id": 90, "LID_evidence": [ { "region_id": 198, "probabilities": { "ja": 0.9822905659675598, "zh": 0.001117385909310542, "en": 0.008897499821614474 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 198 ], "language_confidence": 0.9822905659675598, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1323.4, "packet_end": 1327.3999999999999, "raw_text": "次のご登場をお待ちしております", "selected_text": "次のご利用をお待ちしております", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 209 ], "packet_id": 90, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "次のご登場をお待ちしております", "after": "次のご利用をお待ちしております", "classification": "BEST_SUPPORTED_WITH_REVIEW", "reason": "ASR says 'ご登場' (appearance), but standard phrase is 'ご利用' (use). Annotated 待確認." } ], "language": "ja", "packet_ids": [ 90 ], "index": 209, "independent_ASR": { "text": "次のご登場をお待ちしております。", "chunks": [ { "text": "次のご登場をお待ちしております。", "start": 0.0, "end": 4.0, "chunk_index": 0, "language": "Japanese", "finish_reason": "eos", "truncated": false, "generated_tokens": 8, "max_new_tokens": 128 } ], "detected_language": "Japanese", "segments": [ { "start": 1323.4, "end": 1327.4, "text": "次のご登場をお待ちしております。" } ], "start": 1323.4, "end": 1327.3999999999999, "engine": "mlx-qwen3-asr", "language": "ja" }, "raw_word_timestamps": [ { "word": "次", "start": 1324.18, "end": 1324.5, "probability": 0.99609375 }, { "word": "の", "start": 1324.5, "end": 1324.64, "probability": 1.0 }, { "word": "ご", "start": 1324.64, "end": 1324.76, "probability": 0.99658203125 }, { "word": "登", "start": 1324.76, "end": 1324.92, "probability": 0.8896484375 }, { "word": "場", "start": 1324.92, "end": 1325.1200000000001, "probability": 0.99853515625 }, { "word": "を", "start": 1325.1200000000001, "end": 1325.3200000000002, "probability": 0.57373046875 }, { "word": "お", "start": 1325.3200000000002, "end": 1325.48, "probability": 0.99951171875 }, { "word": "待", "start": 1325.48, "end": 1325.6000000000001, "probability": 1.0 }, { "word": "ち", "start": 1325.6000000000001, "end": 1325.72, "probability": 1.0 }, { "word": "して", "start": 1325.72, "end": 1325.92, "probability": 1.0 }, { "word": "お", "start": 1325.92, "end": 1326.02, "probability": 0.99951171875 }, { "word": "ります", "start": 1326.02, "end": 1326.3200000000002, "probability": 1.0 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "次のご利用をお待ちしております", "review_status": "BEST_SUPPORTED_WITH_REVIEW", "classification": "BEST_SUPPORTED_WITH_REVIEW", "confidence": 0.85, "correction_reason": "ASR says 'ご登場' (appearance), but standard phrase is 'ご利用' (use). Annotated 待確認." }, { "start": 1330.7, "end": 1337.88, "text": "飛行機は間もなく停止いたしますが、安全のためベルト着用サインが消えるまでお座りのままお待ちください。", "raw_asr_text": "飛行機は間もなく停止いたしますが、安全のためベルト着用サインが消えるまでお座りのままお待ちください。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 91, "words": [], "asr_metrics": { "avg_logprob": -0.09362205239229424, "compression_ratio": 1.1811023622047243, "no_speech_prob": 5.641796058108994e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "飛行機は間もなく停止いたしますが、安全のためベルト着用サインが消えるまでお座りのままお待ちください。" } ], "id": 210, "review_annotation": "", "row_id": 210, "bounded_row_id": 210, "parent_row_ids": [ 210 ], "parent_row_id": 210, "packet_id": 91, "LID_evidence": [ { "region_id": 199, "probabilities": { "ja": 0.993465930223465, "zh": 0.000364635662966369, "en": 0.002922904370886196 }, "status": "STABLE" } ], "VAD_region_ids": [ 199 ], "language_confidence": 0.993465930223465, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1329.9, "packet_end": 1339.2, "raw_text": "飛行機は間もなく停止いたしますが、安全のためベルト着用サインが消えるまでお座りのままお待ちください。", "selected_text": "飛行機は間もなく停止いたしますが、安全のためベルト着用サインが消えるまでお座りのままお待ちください。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 210 ], "packet_id": 91, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "飛行機は間もなく停止いたしますが、安全のためベルト着用サインが消えるまでお座りのままお待ちください。", "after": "飛行機は間もなく停止いたしますが、安全のためベルト着用サインが消えるまでお座りのままお待ちください。", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard ANA announcement phrasing matches context and grammar perfectly." } ], "language": "ja", "packet_ids": [ 91 ], "index": 210, "source_text": "飛行機は間もなく停止いたしますが、安全のためベルト着用サインが消えるまでお座りのままお待ちください。", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard ANA announcement phrasing matches context and grammar perfectly.", "raw_word_timestamps": [ { "word": "飛", "start": 1330.48, "end": 1330.8400000000001, "probability": 0.908203125 }, { "word": "行", "start": 1330.8400000000001, "end": 1330.98, "probability": 0.99951171875 }, { "word": "機", "start": 1330.98, "end": 1331.1000000000001, "probability": 0.99755859375 }, { "word": "は", "start": 1331.1000000000001, "end": 1331.26, "probability": 0.99951171875 }, { "word": "間", "start": 1331.26, "end": 1331.42, "probability": 0.60791015625 }, { "word": "も", "start": 1331.42, "end": 1331.5400000000002, "probability": 0.99951171875 }, { "word": "なく", "start": 1331.5400000000002, "end": 1331.8200000000002, "probability": 1.0 }, { "word": "停", "start": 1331.8200000000002, "end": 1332.0800000000002, "probability": 1.0 }, { "word": "止", "start": 1332.0800000000002, "end": 1332.26, "probability": 1.0 }, { "word": "いた", "start": 1332.26, "end": 1332.42, "probability": 0.99267578125 }, { "word": "します", "start": 1332.42, "end": 1332.76, "probability": 1.0 }, { "word": "が、", "start": 1332.76, "end": 1333.0, "probability": 0.99951171875 }, { "word": "安全", "start": 1333.24, "end": 1333.66, "probability": 0.9921875 }, { "word": "の", "start": 1333.66, "end": 1333.8600000000001, "probability": 1.0 }, { "word": "ため", "start": 1333.8600000000001, "end": 1334.1200000000001, "probability": 1.0 }, { "word": "ベ", "start": 1334.1200000000001, "end": 1334.42, "probability": 0.79345703125 }, { "word": "ル", "start": 1334.42, "end": 1334.52, "probability": 1.0 }, { "word": "ト", "start": 1334.52, "end": 1334.6200000000001, "probability": 0.99951171875 }, { "word": "着", "start": 1334.6200000000001, "end": 1334.8000000000002, "probability": 1.0 }, { "word": "用", "start": 1334.8000000000002, "end": 1335.0600000000002, "probability": 1.0 }, { "word": "サ", "start": 1335.0600000000002, "end": 1335.2800000000002, "probability": 0.99560546875 }, { "word": "イ", "start": 1335.2800000000002, "end": 1335.38, "probability": 1.0 }, { "word": "ン", "start": 1335.38, "end": 1335.42, "probability": 1.0 }, { "word": "が", "start": 1335.42, "end": 1335.5400000000002, "probability": 0.99951171875 }, { "word": "消", "start": 1335.5400000000002, "end": 1335.72, "probability": 0.9990234375 }, { "word": "える", "start": 1335.72, "end": 1335.9, "probability": 1.0 }, { "word": "まで", "start": 1335.9, "end": 1336.16, "probability": 0.99951171875 }, { "word": "お", "start": 1336.16, "end": 1336.42, "probability": 0.72802734375 }, { "word": "座", "start": 1336.42, "end": 1336.6000000000001, "probability": 0.998046875 }, { "word": "り", "start": 1336.6000000000001, "end": 1336.7, "probability": 1.0 }, { "word": "の", "start": 1336.7, "end": 1336.8200000000002, "probability": 1.0 }, { "word": "ま", "start": 1336.8200000000002, "end": 1336.96, "probability": 1.0 }, { "word": "ま", "start": 1336.96, "end": 1337.1200000000001, "probability": 1.0 }, { "word": "お", "start": 1337.1200000000001, "end": 1337.3000000000002, "probability": 0.9990234375 }, { "word": "待", "start": 1337.3000000000002, "end": 1337.42, "probability": 1.0 }, { "word": "ち", "start": 1337.42, "end": 1337.52, "probability": 1.0 }, { "word": "ください。", "start": 1337.52, "end": 1337.88, "probability": 0.9990234375 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 1341.9, "end": 1345.0800000000002, "text": "Ladies and gentlemen, the aircraft will come to stop in a moment.", "raw_asr_text": "Ladies and gentlemen, the aircraft will come to stop in a moment.", "source_language": "en", "source_lang": "en", "asr_packet_id": 92, "words": [], "asr_metrics": { "avg_logprob": -0.25536923075831214, "compression_ratio": 1.3174603174603174, "no_speech_prob": 4.193704758459482e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "en", "text": "Ladies and gentlemen, the aircraft will come to stop in a moment. For your own safety, please you may seated until the captain turns off the seatbelt sign. Thank you." } ], "id": 211, "review_annotation": "", "row_id": 211, "bounded_row_id": 211, "parent_row_ids": [ 211 ], "parent_row_id": 211, "packet_id": 92, "LID_evidence": [ { "region_id": 200, "probabilities": { "ja": 0.001299533192650415, "zh": 0.004826027448871173, "en": 0.9772800356149673 }, "status": "STABLE" } ], "VAD_region_ids": [ 200 ], "language_confidence": 0.9772800356149673, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1341.1000000000001, "packet_end": 1351.8, "raw_text": "Ladies and gentlemen, the aircraft will come to stop in a moment.", "selected_text": "Ladies and gentlemen, the aircraft will come to stop in a moment.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 211 ], "packet_id": 92, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "Ladies and gentlemen, the aircraft will come to stop in a moment.", "after": "Ladies and gentlemen, the aircraft will come to stop in a moment.", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Grammatically correct standard announcement text supported by context." } ], "language": "en", "packet_ids": [ 92 ], "index": 211, "source_text": "Ladies and gentlemen, the aircraft will come to stop in a moment.", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.95, "correction_reason": "Grammatically correct standard announcement text supported by context.", "raw_word_timestamps": [ { "word": " Ladies", "start": 1341.7400000000002, "end": 1342.22, "probability": 0.56591796875 }, { "word": " and", "start": 1342.22, "end": 1342.5200000000002, "probability": 0.98681640625 }, { "word": " gentlemen,", "start": 1342.5200000000002, "end": 1342.8400000000001, "probability": 0.80615234375 }, { "word": " the", "start": 1343.0400000000002, "end": 1343.1000000000001, "probability": 0.98388671875 }, { "word": " aircraft", "start": 1343.1000000000001, "end": 1343.3400000000001, "probability": 0.56005859375 }, { "word": " will", "start": 1343.3400000000001, "end": 1343.7400000000002, "probability": 0.94189453125 }, { "word": " come", "start": 1343.7400000000002, "end": 1343.94, "probability": 0.9951171875 }, { "word": " to", "start": 1343.94, "end": 1344.1200000000001, "probability": 0.986328125 }, { "word": " stop", "start": 1344.1200000000001, "end": 1344.42, "probability": 0.87109375 }, { "word": " in", "start": 1344.42, "end": 1344.6000000000001, "probability": 0.9228515625 }, { "word": " a", "start": 1344.6000000000001, "end": 1344.72, "probability": 0.99267578125 }, { "word": " moment.", "start": 1344.72, "end": 1345.0800000000002, "probability": 0.9990234375 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 1345.5400000000002, "end": 1349.9800000000002, "text": "For your own safety, please remain seated until the captain turns off the seatbelt sign.", "raw_asr_text": "For your own safety, please you may seated until the captain turns off the seatbelt sign.", "source_language": "en", "source_lang": "en", "asr_packet_id": 92, "words": [], "asr_metrics": { "avg_logprob": -0.25536923075831214, "compression_ratio": 1.3174603174603174, "no_speech_prob": 4.193704758459482e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "en", "text": "Ladies and gentlemen, the aircraft will come to stop in a moment. For your own safety, please you may seated until the captain turns off the seatbelt sign. Thank you." } ], "id": 212, "review_annotation": "", "row_id": 212, "bounded_row_id": 212, "parent_row_ids": [ 212 ], "parent_row_id": 212, "packet_id": 92, "LID_evidence": [ { "region_id": 200, "probabilities": { "ja": 0.001299533192650415, "zh": 0.004826027448871173, "en": 0.9772800356149673 }, "status": "STABLE" }, { "region_id": 201, "probabilities": { "ja": 0.19137832355142262, "zh": 0.004516460141167045, "en": 0.774850478162989 }, "status": "STABLE" } ], "VAD_region_ids": [ 200, 201 ], "language_confidence": 0.9772800356149673, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1341.1000000000001, "packet_end": 1351.8, "raw_text": "For your own safety, please you may seated until the captain turns off the seatbelt sign.", "selected_text": "For your own safety, please remain seated until the captain turns off the seatbelt sign.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 212 ], "packet_id": 92, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "For your own safety, please you may seated until the captain turns off the seatbelt sign.", "after": "For your own safety, please remain seated until the captain turns off the seatbelt sign.", "classification": "CERTAIN_RESTORATION", "reason": "Independent ASR corrects 'you may' to 'remain', which is grammatically correct and standard." } ], "language": "en", "packet_ids": [ 92 ], "index": 212, "independent_ASR": { "text": "Ladies and gentlemen, the aircraft will come to stop in a moment. For your own safety, please remain seated until the captain turns off the seatbelt sign. Thank you.", "chunks": [ { "text": "Ladies and gentlemen, the aircraft will come to stop in a moment. For your own safety, please remain seated until the captain turns off the seatbelt sign. Thank you.", "start": 0.0, "end": 10.7, "chunk_index": 0, "language": "English", "finish_reason": "eos", "truncated": false, "generated_tokens": 36, "max_new_tokens": 161 } ], "detected_language": "English", "segments": [ { "start": 1341.1000000000001, "end": 1351.8000000000002, "text": "Ladies and gentlemen, the aircraft will come to stop in a moment. For your own safety, please remain seated until the captain turns off the seatbelt sign. Thank you." } ], "start": 1341.1000000000001, "end": 1351.8, "engine": "mlx-qwen3-asr", "language": "en" }, "raw_word_timestamps": [ { "word": " For", "start": 1345.5400000000002, "end": 1346.0200000000002, "probability": 0.98486328125 }, { "word": " your", "start": 1346.0200000000002, "end": 1346.2, "probability": 0.95068359375 }, { "word": " own", "start": 1346.2, "end": 1346.3600000000001, "probability": 0.740234375 }, { "word": " safety,", "start": 1346.3600000000001, "end": 1346.7400000000002, "probability": 0.998046875 }, { "word": " please", "start": 1346.92, "end": 1347.0600000000002, "probability": 0.99169921875 }, { "word": " you", "start": 1347.0600000000002, "end": 1347.2600000000002, "probability": 0.360107421875 }, { "word": " may", "start": 1347.2600000000002, "end": 1347.3400000000001, "probability": 0.9951171875 }, { "word": " seated", "start": 1347.3400000000001, "end": 1347.72, "probability": 0.57373046875 }, { "word": " until", "start": 1347.72, "end": 1348.14, "probability": 0.8203125 }, { "word": " the", "start": 1348.14, "end": 1348.3200000000002, "probability": 0.9853515625 }, { "word": " captain", "start": 1348.3200000000002, "end": 1348.5800000000002, "probability": 0.8671875 }, { "word": " turns", "start": 1348.5800000000002, "end": 1348.92, "probability": 0.9921875 }, { "word": " off", "start": 1348.92, "end": 1349.1000000000001, "probability": 0.951171875 }, { "word": " the", "start": 1349.1000000000001, "end": 1349.2400000000002, "probability": 0.98828125 }, { "word": " seatbelt", "start": 1349.2400000000002, "end": 1349.6200000000001, "probability": 0.931640625 }, { "word": " sign.", "start": 1349.6200000000001, "end": 1349.9800000000002, "probability": 0.99267578125 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "For your own safety, please remain seated until the captain turns off the seatbelt sign.", "review_status": "CERTAIN_RESTORATION", "classification": "CERTAIN_RESTORATION", "confidence": 0.95, "correction_reason": "Independent ASR corrects 'you may' to 'remain', which is grammatically correct and standard." }, { "start": 1350.18, "end": 1350.5600000000002, "text": "Thank you.", "raw_asr_text": "Thank you.", "source_language": "en", "source_lang": "en", "asr_packet_id": 92, "words": [ { "word": " Thank", "start": 1350.18, "end": 1350.3200000000002, "probability": 0.990234375 }, { "word": " you.", "start": 1350.3200000000002, "end": 1350.5600000000002, "probability": 0.98779296875 } ], "asr_metrics": { "avg_logprob": -0.25536923075831214, "compression_ratio": 1.3174603174603174, "no_speech_prob": 4.193704758459482e-11 }, "lid_status": "STABLE", "alternatives": [ { "language": "en", "text": "Ladies and gentlemen, the aircraft will come to stop in a moment. For your own safety, please you may seated until the captain turns off the seatbelt sign. Thank you." } ], "id": 213, "review_annotation": "", "row_id": 213, "bounded_row_id": 213, "parent_row_ids": [ 213 ], "parent_row_id": 213, "packet_id": 92, "LID_evidence": [ { "region_id": 201, "probabilities": { "ja": 0.19137832355142262, "zh": 0.004516460141167045, "en": 0.774850478162989 }, "status": "STABLE" } ], "VAD_region_ids": [ 201 ], "language_confidence": 0.774850478162989, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1341.1000000000001, "packet_end": 1351.8, "raw_text": "Thank you.", "selected_text": "Thank you.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 213 ], "packet_id": 92, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "Thank you.", "after": "Thank you.", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard closing phrase, clearly distinct and supported by context." } ], "language": "en", "packet_ids": [ 92 ], "index": 213, "source_text": "Thank you.", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.99, "correction_reason": "Standard closing phrase, clearly distinct and supported by context." }, { "start": 1351.54, "end": 1356.78, "text": "皆様、機内に忘れ物のないよう、上の物入れ、座席ポケットをお確かめください。", "raw_asr_text": "皆様、機内に忘れ物のないよう、上の物入れ、座席ポケットをお確かめください。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 93, "words": [ { "word": "皆", "start": 1351.54, "end": 1351.8600000000001, "probability": 0.86572265625 }, { "word": "様、", "start": 1351.8600000000001, "end": 1352.1200000000001, "probability": 0.79296875 }, { "word": "機", "start": 1352.26, "end": 1352.3400000000001, "probability": 0.98486328125 }, { "word": "内", "start": 1352.3400000000001, "end": 1352.48, "probability": 0.9990234375 }, { "word": "に", "start": 1352.48, "end": 1352.64, "probability": 0.9990234375 }, { "word": "忘", "start": 1352.64, "end": 1352.8600000000001, "probability": 1.0 }, { "word": "れ", "start": 1352.8600000000001, "end": 1353.0, "probability": 1.0 }, { "word": "物", "start": 1353.0, "end": 1353.18, "probability": 0.998046875 }, { "word": "の", "start": 1353.18, "end": 1353.3600000000001, "probability": 0.9912109375 }, { "word": "ない", "start": 1353.3600000000001, "end": 1353.5800000000002, "probability": 0.9931640625 }, { "word": "よう、", "start": 1353.5800000000002, "end": 1353.76, "probability": 0.97802734375 }, { "word": "上", "start": 1353.82, "end": 1353.96, "probability": 0.97802734375 }, { "word": "の", "start": 1353.96, "end": 1354.1000000000001, "probability": 0.99462890625 }, { "word": "物", "start": 1354.1000000000001, "end": 1354.3, "probability": 0.890625 }, { "word": "入", "start": 1354.3, "end": 1354.5, "probability": 0.94287109375 }, { "word": "れ、", "start": 1354.5, "end": 1354.6200000000001, "probability": 0.99609375 }, { "word": "座", "start": 1354.66, "end": 1354.78, "probability": 0.998046875 }, { "word": "席", "start": 1354.78, "end": 1355.0, "probability": 1.0 }, { "word": "ポ", "start": 1355.0, "end": 1355.18, "probability": 0.96044921875 }, { "word": "ケ", "start": 1355.18, "end": 1355.3400000000001, "probability": 1.0 }, { "word": "ット", "start": 1355.3400000000001, "end": 1355.56, "probability": 1.0 }, { "word": "を", "start": 1355.56, "end": 1355.72, "probability": 0.9970703125 }, { "word": "お", "start": 1355.72, "end": 1355.8400000000001, "probability": 0.9921875 }, { "word": "確", "start": 1355.8400000000001, "end": 1356.0, "probability": 0.99951171875 }, { "word": "か", "start": 1356.0, "end": 1356.18, "probability": 0.9990234375 }, { "word": "め", "start": 1356.18, "end": 1356.32, "probability": 0.99951171875 }, { "word": "ください。", "start": 1356.32, "end": 1356.78, "probability": 0.99853515625 } ], "asr_metrics": { "avg_logprob": -0.11464820924352427, "compression_ratio": 1.3161290322580645, "no_speech_prob": 6.941020241207907e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "皆様、機内に忘れ物のないよう、上の物入れ、座席ポケットをお確かめください。外は冷たい雨が降っております。どうぞ気をつけてお帰りください。" } ], "id": 214, "review_annotation": "", "row_id": 214, "bounded_row_id": 214, "parent_row_ids": [ 214 ], "parent_row_id": 214, "packet_id": 93, "LID_evidence": [ { "region_id": 202, "probabilities": { "ja": 0.9849929004907608, "zh": 0.000913574117657845, "en": 0.006574819891829975 }, "status": "STABLE" } ], "VAD_region_ids": [ 202 ], "language_confidence": 0.9849929004907608, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1350.7, "packet_end": 1362.3999999999999, "raw_text": "皆様、機内に忘れ物のないよう、上の物入れ、座席ポケットをお確かめください。", "selected_text": "皆様、機内に忘れ物のないよう、上の物入れ、座席ポケットをお確かめください。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 214 ], "packet_id": 93, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "皆様、機内に忘れ物のないよう、上の物入れ、座席ポケットをお確かめください。", "after": "皆様、機内に忘れ物のないよう、上の物入れ、座席ポケットをお確かめください。", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard ANA announcement phrasing matches context and grammar perfectly." } ], "language": "ja", "packet_ids": [ 93 ], "index": 214, "source_text": "皆様、機内に忘れ物のないよう、上の物入れ、座席ポケットをお確かめください。", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard ANA announcement phrasing matches context and grammar perfectly." }, { "start": 1357.38, "end": 1361.0800000000002, "text": "外は冷たい雨が降っております。どうぞ気をつけてお帰りください。", "raw_asr_text": "外は冷たい雨が降っております。どうぞ気をつけてお帰りください。", "source_language": "ja", "source_lang": "ja", "asr_packet_id": 93, "words": [ { "word": "外", "start": 1357.38, "end": 1357.54, "probability": 0.990234375 }, { "word": "は", "start": 1357.54, "end": 1357.72, "probability": 0.99951171875 }, { "word": "冷", "start": 1357.72, "end": 1357.92, "probability": 0.99658203125 }, { "word": "たい", "start": 1357.92, "end": 1358.18, "probability": 0.98828125 }, { "word": "雨", "start": 1358.18, "end": 1358.32, "probability": 1.0 }, { "word": "が", "start": 1358.32, "end": 1358.5, "probability": 1.0 }, { "word": "降", "start": 1358.5, "end": 1358.64, "probability": 0.99951171875 }, { "word": "って", "start": 1358.64, "end": 1358.8, "probability": 1.0 }, { "word": "お", "start": 1358.8, "end": 1358.9, "probability": 0.998046875 }, { "word": "ります。", "start": 1358.9, "end": 1359.16, "probability": 0.99951171875 }, { "word": "どう", "start": 1359.26, "end": 1359.32, "probability": 0.99609375 }, { "word": "ぞ", "start": 1359.32, "end": 1359.44, "probability": 0.99951171875 }, { "word": "気", "start": 1359.44, "end": 1359.64, "probability": 0.5302734375 }, { "word": "を", "start": 1359.64, "end": 1359.78, "probability": 1.0 }, { "word": "つ", "start": 1359.78, "end": 1359.8600000000001, "probability": 0.95751953125 }, { "word": "けて", "start": 1359.8600000000001, "end": 1360.1000000000001, "probability": 1.0 }, { "word": "お", "start": 1360.1000000000001, "end": 1360.3, "probability": 0.9990234375 }, { "word": "帰", "start": 1360.3, "end": 1360.52, "probability": 0.998046875 }, { "word": "り", "start": 1360.52, "end": 1360.64, "probability": 0.99951171875 }, { "word": "ください。", "start": 1360.64, "end": 1361.0800000000002, "probability": 0.99755859375 } ], "asr_metrics": { "avg_logprob": -0.11464820924352427, "compression_ratio": 1.3161290322580645, "no_speech_prob": 6.941020241207907e-12 }, "lid_status": "STABLE", "alternatives": [ { "language": "ja", "text": "皆様、機内に忘れ物のないよう、上の物入れ、座席ポケットをお確かめください。外は冷たい雨が降っております。どうぞ気をつけてお帰りください。" } ], "id": 215, "review_annotation": "", "row_id": 215, "bounded_row_id": 215, "parent_row_ids": [ 215 ], "parent_row_id": 215, "packet_id": 93, "LID_evidence": [ { "region_id": 202, "probabilities": { "ja": 0.9849929004907608, "zh": 0.000913574117657845, "en": 0.006574819891829975 }, "status": "STABLE" } ], "VAD_region_ids": [ 202 ], "language_confidence": 0.9849929004907608, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1350.7, "packet_end": 1362.3999999999999, "raw_text": "外は冷たい雨が降っております。どうぞ気をつけてお帰りください。", "selected_text": "外は冷たい雨が降っております。どうぞ気をつけてお帰りください。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 215 ], "packet_id": 93, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "外は冷たい雨が降っております。どうぞ気をつけてお帰りください。", "after": "外は冷たい雨が降っております。どうぞ気をつけてお帰りください。", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard closing announcement phrasing matches context and grammar perfectly." } ], "language": "ja", "packet_ids": [ 93 ], "index": 215, "source_text": "外は冷たい雨が降っております。どうぞ気をつけてお帰りください。", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.98, "correction_reason": "Standard closing announcement phrasing matches context and grammar perfectly." }, { "start": 1366.0, "end": 1367.9, "text": "ありがとうございました。", "raw_asr_text": "", "source_language": "ja", "source_lang": "ja", "vad_region_id": 203, "review_status": "CERTAIN_RESTORATION", "asr_packet_id": 94, "correction_reason": "Independent ASR provides clear 'Thank you' matching the English segment's closing.", "confidence": 0.95, "id": 216, "review_annotation": "", "row_id": 216, "bounded_row_id": 216, "parent_row_ids": [ 216 ], "parent_row_id": 216, "packet_id": 94, "LID_evidence": [ { "region_id": 203, "probabilities": { "ja": 0, "zh": 0, "en": 0 }, "status": "UNCERTAIN" } ], "VAD_region_ids": [ 203 ], "language_confidence": 0, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1365.2, "packet_end": 1368.7, "raw_text": "", "selected_text": "ありがとうございました。", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 216 ], "packet_id": 94, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "[聴不清／待確認]", "after": "ありがとうございました。", "classification": "CERTAIN_RESTORATION", "reason": "Independent ASR provides clear 'Thank you' matching the English segment's closing." } ], "language": "ja", "packet_ids": [ 94 ], "index": 216, "independent_ASR": { "text": "ありがとうございました。", "chunks": [ { "text": "ありがとうございました。", "start": 0.0, "end": 3.5, "chunk_index": 0, "language": "Japanese", "finish_reason": "eos", "truncated": false, "generated_tokens": 2, "max_new_tokens": 128 } ], "detected_language": "Japanese", "segments": [ { "start": 1365.2, "end": 1368.7, "text": "ありがとうございました。" } ], "start": 1365.2, "end": 1368.7, "engine": "mlx-qwen3-asr", "language": "ja" }, "raw_word_timestamps": [], "words": [], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "ありがとうございました。", "classification": "CERTAIN_RESTORATION" }, { "start": 1374.3, "end": 1378.02, "text": "Ladies and gentlemen, please make sure to take all your belongings with you when you", "raw_asr_text": "Ladies and gentlemen, please make sure to take all your belongings with you when you", "source_language": "en", "source_lang": "en", "asr_packet_id": 95, "words": [], "asr_metrics": { "avg_logprob": -0.4451215020541487, "compression_ratio": 1.1910112359550562, "no_speech_prob": 1.6704047867133198e-10 }, "lid_status": "STABLE", "alternatives": [ { "language": "en", "text": "Ladies and gentlemen, please make sure to take all your belongings with you when you deep lane. Thank you." } ], "id": 217, "review_annotation": "", "row_id": 217, "bounded_row_id": 217, "parent_row_ids": [ 217 ], "parent_row_id": 217, "packet_id": 95, "LID_evidence": [ { "region_id": 204, "probabilities": { "ja": 0.004514704724507672, "zh": 0.0032095228754250066, "en": 0.9167507801737104 }, "status": "STABLE" } ], "VAD_region_ids": [ 204 ], "language_confidence": 0.9167507801737104, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1373.5, "packet_end": 1380.3, "raw_text": "Ladies and gentlemen, please make sure to take all your belongings with you when you", "selected_text": "Ladies and gentlemen, please make sure to take all your belongings with you when you", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 217 ], "packet_id": 95, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "Ladies and gentlemen, please make sure to take all your belongings with you when you", "after": "Ladies and gentlemen, please make sure to take all your belongings with you when you", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Raw ASR is incomplete but matches the start of the independent ASR segment." } ], "language": "en", "packet_ids": [ 95 ], "index": 217, "independent_ASR": { "text": "Ladies and gentlemen, please make sure to take all your belongings with you when you depart. Thank you.", "chunks": [ { "text": "Ladies and gentlemen, please make sure to take all your belongings with you when you depart. Thank you.", "start": 0.0, "end": 6.8, "chunk_index": 0, "language": "English", "finish_reason": "eos", "truncated": false, "generated_tokens": 22, "max_new_tokens": 128 } ], "detected_language": "English", "segments": [ { "start": 1373.5, "end": 1380.3, "text": "Ladies and gentlemen, please make sure to take all your belongings with you when you depart. Thank you." } ], "start": 1373.5, "end": 1380.3, "engine": "mlx-qwen3-asr", "language": "en" }, "source_text": "Ladies and gentlemen, please make sure to take all your belongings with you when you", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.9, "correction_reason": "Raw ASR is incomplete but matches the start of the independent ASR segment.", "raw_word_timestamps": [ { "word": " Ladies", "start": 1374.18, "end": 1374.66, "probability": 0.5693359375 }, { "word": " and", "start": 1374.66, "end": 1374.94, "probability": 0.9765625 }, { "word": " gentlemen,", "start": 1374.94, "end": 1375.22, "probability": 0.87255859375 }, { "word": " please", "start": 1375.42, "end": 1375.58, "probability": 0.97265625 }, { "word": " make", "start": 1375.58, "end": 1375.82, "probability": 0.9736328125 }, { "word": " sure", "start": 1375.82, "end": 1376.04, "probability": 0.998046875 }, { "word": " to", "start": 1376.04, "end": 1376.22, "probability": 0.98974609375 }, { "word": " take", "start": 1376.22, "end": 1376.46, "probability": 0.99072265625 }, { "word": " all", "start": 1376.46, "end": 1376.7, "probability": 0.98486328125 }, { "word": " your", "start": 1376.7, "end": 1376.84, "probability": 0.9638671875 }, { "word": " belongings", "start": 1376.84, "end": 1377.24, "probability": 0.9970703125 }, { "word": " with", "start": 1377.24, "end": 1377.56, "probability": 0.9033203125 }, { "word": " you", "start": 1377.56, "end": 1377.66, "probability": 0.98974609375 }, { "word": " when", "start": 1377.66, "end": 1377.9, "probability": 0.9033203125 }, { "word": " you", "start": 1377.9, "end": 1378.02, "probability": 0.97216796875 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window; raw word alignment unusable" }, { "start": 1378.02, "end": 1378.52, "text": "depart. Thank you.", "raw_asr_text": "deep lane.", "source_language": "en", "source_lang": "en", "asr_packet_id": 95, "words": [], "asr_metrics": { "avg_logprob": -0.4451215020541487, "compression_ratio": 1.1910112359550562, "no_speech_prob": 1.6704047867133198e-10 }, "lid_status": "STABLE", "alternatives": [ { "language": "en", "text": "Ladies and gentlemen, please make sure to take all your belongings with you when you deep lane. Thank you." } ], "id": 218, "review_annotation": "", "row_id": 218, "bounded_row_id": 218, "parent_row_ids": [ 218 ], "parent_row_id": 218, "packet_id": 95, "LID_evidence": [ { "region_id": 204, "probabilities": { "ja": 0.004514704724507672, "zh": 0.0032095228754250066, "en": 0.9167507801737104 }, "status": "STABLE" } ], "VAD_region_ids": [ 204 ], "language_confidence": 0.9167507801737104, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1373.5, "packet_end": 1380.3, "raw_text": "deep lane.", "selected_text": "depart. Thank you.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 218 ], "packet_id": 95, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "deep lane.", "after": "depart. Thank you.", "classification": "CERTAIN_RESTORATION", "reason": "Independent ASR confirms 'depart' replacing nonsense 'deep lane'." } ], "language": "en", "packet_ids": [ 95 ], "index": 218, "independent_ASR": { "text": "Ladies and gentlemen, please make sure to take all your belongings with you when you depart. Thank you.", "chunks": [ { "text": "Ladies and gentlemen, please make sure to take all your belongings with you when you depart. Thank you.", "start": 0.0, "end": 6.8, "chunk_index": 0, "language": "English", "finish_reason": "eos", "truncated": false, "generated_tokens": 22, "max_new_tokens": 128 } ], "detected_language": "English", "segments": [ { "start": 1373.5, "end": 1380.3, "text": "Ladies and gentlemen, please make sure to take all your belongings with you when you depart. Thank you." } ], "start": 1373.5, "end": 1380.3, "engine": "mlx-qwen3-asr", "language": "en" }, "raw_word_timestamps": [ { "word": " deep", "start": 1378.02, "end": 1378.24, "probability": 0.068359375 }, { "word": " lane.", "start": 1378.24, "end": 1378.52, "probability": 0.4453125 } ], "word_timestamps": [], "canonical_timing_evidence": "conservative_parent_cue_window", "source_text": "depart. Thank you.", "review_status": "CERTAIN_RESTORATION", "classification": "CERTAIN_RESTORATION", "confidence": 0.95, "correction_reason": "Independent ASR confirms 'depart' replacing nonsense 'deep lane'." }, { "start": 1378.68, "end": 1379.06, "text": "Thank you.", "raw_asr_text": "Thank you.", "source_language": "en", "source_lang": "en", "asr_packet_id": 95, "words": [ { "word": " Thank", "start": 1378.68, "end": 1378.76, "probability": 0.99365234375 }, { "word": " you.", "start": 1378.76, "end": 1379.06, "probability": 0.984375 } ], "asr_metrics": { "avg_logprob": -0.4451215020541487, "compression_ratio": 1.1910112359550562, "no_speech_prob": 1.6704047867133198e-10 }, "lid_status": "STABLE", "alternatives": [ { "language": "en", "text": "Ladies and gentlemen, please make sure to take all your belongings with you when you deep lane. Thank you." } ], "id": 219, "review_annotation": "", "row_id": 219, "bounded_row_id": 219, "parent_row_ids": [ 219 ], "parent_row_id": 219, "packet_id": 95, "LID_evidence": [ { "region_id": 204, "probabilities": { "ja": 0.004514704724507672, "zh": 0.0032095228754250066, "en": 0.9167507801737104 }, "status": "STABLE" } ], "VAD_region_ids": [ 204 ], "language_confidence": 0.9167507801737104, "ASR_engine": "mlx-whisper", "ASR_route": "bounded_vad_lid", "packet_start": 1373.5, "packet_end": 1380.3, "raw_text": "Thank you.", "selected_text": "Thank you.", "provenance_chain": [ { "stage": "bounded_asr", "parent_row_ids": [ 219 ], "packet_id": 95, "engine": "mlx-whisper" }, { "stage": "canonical_review", "before": "Thank you.", "after": "Thank you.", "classification": "VALID_SPOKEN_LANGUAGE", "reason": "Standard closing phrase, clearly distinct and supported by context." } ], "language": "en", "packet_ids": [ 95 ], "index": 219, "source_text": "Thank you.", "review_status": "VALID_SPOKEN_LANGUAGE", "classification": "VALID_SPOKEN_LANGUAGE", "confidence": 0.99, "correction_reason": "Standard closing phrase, clearly distinct and supported by context." } ], "transcript_selection": { "schema": "videonote.evidence-preserving-selector.v1", "selected_source": "bounded_multilingual_asr_local_review", "whisper_rejected": false, "source_profile": { "duration_coverage": null }, "decision_reason": "bounded_backbone_preserved_local_QC", "handoff_report": "/Users/sagawa/AI_WORK/video_notes/url/20261012_063243__160314_ANA852便_台北松山_羽田_機内アナウンス集/production_selector_handoff_report.json" }, "canonical_stage": "complete_production_evidence_review", "presegmentation_correction": { "applied": false, "correction_count": 0, "deterministic_fix_count": 0 } }