{
  "text": "各位小夥伴，大家好。我是瘋一樣，今天給大家帶來NextSub 5.0版本更新。\n從5.0開始，軟件名稱正式從NextSub改稿，改為NextSub，Logo也從青蛙換成閃電。\n因為我們是離線字幕工具,為了減少版本更新給大家帶來的麻煩,後續我們不會優先開發新功能,而是對5.0版本做長期維護,大家可以把5.0當作穩定主力版本長期使用。\nNextSub 5.0新增支持Premiere 2022-2026版本的一鍵拉取字幕，字幕智能分段，字幕智能轉換，字幕處理完成後，導入剪輯軟件，還支持自動標記。\n想要使用Premiere一鍵拉取功能，需要先安裝Premiere橋接插件，直接雙擊Back文件，就能完成安裝。\n裝好之後,同時打開Premiere和NextSub,軟件會自動檢測並完成橋接,連接狀態可以在日志裡查看。\n當我們字幕中有大量中文數字或者物理數學符號的時候,做字幕的時候需要轉換成阿拉伯數字和圖形符號。\n字幕智能轉換功能可以自動把字幕文本裡的中文數字轉為阿拉伯數字,不用每條字幕手動修改。\n轉換時會彈出轉換報告,方便核對。\n最後,當我們把處理好的字幕導入剪輯軟件時,所有經過轉換的字幕會自動打上標記,方便快速檢查。\n針對超長字幕,我們開發了字幕智能分段功能,它會按照你設置的字符數量自動拆分,完成後彈出分割報告,方便核對。\n以上就是本次更新的核心內容。\n大家遇到任何問題,都可以通過QQ群反饋。\n感謝觀看,我們下次再見。",
  "language": "zh",
  "duration": 120.768,
  "segments": [
    {
      "id": 0,
      "seek": 0,
      "start": 0.0,
      "end": 6.9,
      "text": "各位小夥伴，大家好。我是瘋一樣，今天給大家帶來NextSub 5.0版本更新。",
      "tokens": [
        50365,
        22073,
        7322,
        1787,
        98,
        7384,
        112,
        11,
        15248,
        1543,
        15914,
        12760,
        163,
        246,
        233,
        22977,
        11,
        12074,
        17798,
        6868,
        32379,
        3763,
        31002,
        39582,
        1025,
        13,
        15,
        42096,
        8802,
        19002,
        12560,
        1543,
        50710
      ],
      "temperature": 0,
      "avg_logprob": -0.12159411112467448,
      "compression_ratio": 1.273615635179153,
      "no_speech_prob": 9.256976261917682e-12,
      "source": "whisper",
      "text_original": "各位小夥伴,大家好。我是像瘋一樣,今天給大家帶來NextSub 5.0版本更新。",
      "correction_applied": true,
      "correction_source": "llm_correct_a_presegmentation"
    },
    {
      "id": 1,
      "seek": 0,
      "start": 7.5,
      "end": 14.3,
      "text": "從5.0開始，軟件名稱正式從NextSub改稿，改為NextSub，Logo也從青蛙換成閃電。",
      "tokens": [
        50740,
        21068,
        20,
        13,
        15,
        21017,
        11,
        14454,
        253,
        20485,
        15940,
        49606,
        15789,
        27584,
        21068,
        10415,
        123,
        11,
        34490,
        6344,
        31002,
        39582,
        11,
        43,
        23515,
        6404,
        21068,
        37462,
        26145,
        247,
        36338,
        11336,
        5196,
        225,
        20545,
        1543,
        51080
      ],
      "temperature": 0,
      "avg_logprob": -0.12159411112467448,
      "compression_ratio": 1.273615635179153,
      "no_speech_prob": 9.256976261917682e-12,
      "source": "whisper",
      "text_original": "從5.0開始,軟件名稱正式從稿,改為NextSub,Logo也從青蛙換成閃電。",
      "correction_applied": true,
      "correction_source": "llm_correct_a_presegmentation"
    },
    {
      "id": 2,
      "seek": 0,
      "start": 14.92,
      "end": 28.3,
      "text": "因為我們是離線字幕工具,為了減少版本更新給大家帶來的麻煩,後續我們不會優先開發新功能,而是對5.0版本做長期維護,大家可以把5.0當作穩定主力版本長期使用。",
      "tokens": [
        51111,
        11471,
        5884,
        1541,
        32662,
        31699,
        22381,
        40211,
        23323,
        39806,
        11,
        6344,
        2289,
        9592,
        249,
        15686,
        42096,
        8802,
        19002,
        12560,
        17798,
        6868,
        32379,
        3763,
        1546,
        38999,
        17867,
        102,
        11,
        5661,
        29583,
        5884,
        21121,
        12827,
        103,
        10108,
        8949,
        14637,
        12560,
        32311,
        8225,
        11,
        11070,
        1541,
        2855,
        20,
        13,
        15,
        42096,
        8802,
        10907,
        15353,
        16786,
        9261,
        255,
        48289,
        11,
        6868,
        6723,
        16075,
        20,
        13,
        15,
        13118,
        11914,
        13740,
        102,
        12088,
        13557,
        13486,
        42096,
        8802,
        15353,
        16786,
        22982,
        9254,
        1543,
        51780
      ],
      "temperature": 0,
      "avg_logprob": -0.12159411112467448,
      "compression_ratio": 1.273615635179153,
      "no_speech_prob": 9.256976261917682e-12,
      "source": "whisper"
    },
    {
      "id": 3,
      "seek": 2830,
      "start": 28.3,
      "end": 41.66,
      "text": "NextSub 5.0新增支持Premiere 2022-2026版本的一鍵拉取字幕，字幕智能分段，字幕智能轉換，字幕處理完成後，導入剪輯軟件，還支持自動標記。",
      "tokens": [
        50365,
        31002,
        39582,
        1025,
        13,
        15,
        12560,
        24228,
        252,
        35488,
        47,
        2579,
        14412,
        20229,
        12,
        2009,
        10880,
        42096,
        8802,
        1546,
        2257,
        41668,
        113,
        29717,
        29436,
        22381,
        40211,
        11,
        22381,
        40211,
        5094,
        118,
        8225,
        6627,
        28427,
        11,
        22381,
        40211,
        5094,
        118,
        8225,
        32739,
        36338,
        11,
        22381,
        40211,
        39289,
        13876,
        41509,
        5661,
        11,
        25647,
        14028,
        5935,
        103,
        12290,
        107,
        14454,
        253,
        20485,
        11,
        7824,
        35488,
        9722,
        12572,
        46299,
        16958,
        1543,
        51033
      ],
      "temperature": 0,
      "avg_logprob": -0.09875002400628452,
      "compression_ratio": 1.1733870967741935,
      "no_speech_prob": 1.1713739353491626e-11,
      "source": "whisper",
      "text_original": "NextSub 5.0新增支持Premiere 2022-2026版本的一鍵拉取字幕,字幕智能分段,字幕智能轉換,字幕處理完成後,導入剪輯軟件,還支持自動標記。",
      "correction_applied": true,
      "correction_source": "llm_correct_a_presegmentation"
    },
    {
      "id": 4,
      "seek": 2830,
      "start": 42.0,
      "end": 50.06,
      "text": "想要使用Premiere一鍵拉取功能，需要先安裝Premiere橋接插件，直接雙擊Back文件，就能完成安裝。",
      "tokens": [
        51050,
        7093,
        4275,
        22982,
        9254,
        47,
        2579,
        14412,
        2257,
        41668,
        113,
        29717,
        29436,
        32311,
        8225,
        11,
        35748,
        10108,
        16206,
        44453,
        47,
        2579,
        14412,
        21416,
        98,
        14468,
        11673,
        240,
        20485,
        11,
        43297,
        6306,
        247,
        37813,
        28404,
        17174,
        20485,
        11,
        3111,
        8225,
        41509,
        16206,
        44453,
        1543,
        51453
      ],
      "temperature": 0,
      "avg_logprob": -0.09875002400628452,
      "compression_ratio": 1.1733870967741935,
      "no_speech_prob": 1.1713739353491626e-11,
      "source": "whisper",
      "text_original": "想要使用Premiere一鍵拉取功能,需要先安裝Premiere桥接插件,直接雙擊Back文件,就能完成安裝。",
      "correction_applied": true,
      "correction_source": "llm_correct_a_presegmentation"
    },
    {
      "id": 5,
      "seek": 5006,
      "start": 50.06,
      "end": 59.36,
      "text": "裝好之後,同時打開Premiere和NextSub,軟件會自動檢測並完成橋接,連接狀態可以在日志裡查看。",
      "tokens": [
        50365,
        44453,
        2131,
        26362,
        11,
        13089,
        6611,
        12467,
        8949,
        47,
        2579,
        14412,
        12565,
        31002,
        39582,
        11,
        14454,
        253,
        20485,
        6236,
        9722,
        12572,
        40741,
        95,
        9592,
        105,
        35934,
        41509,
        36473,
        14468,
        11,
        36252,
        14468,
        18637,
        222,
        45248,
        6723,
        3581,
        6890,
        44700,
        11066,
        42623,
        4200,
        1543,
        50830
      ],
      "temperature": 0,
      "avg_logprob": -0.07672803119946552,
      "compression_ratio": 1.1163793103448276,
      "no_speech_prob": 1.9141077611806168e-11,
      "source": "whisper"
    },
    {
      "id": 6,
      "seek": 5006,
      "start": 60.040000000000006,
      "end": 68.2,
      "text": "當我們字幕中有大量中文數字或者物理數學符號的時候,做字幕的時候需要轉換成阿拉伯數字和圖形符號。",
      "tokens": [
        50864,
        13118,
        5884,
        22381,
        40211,
        5975,
        2412,
        3582,
        26748,
        5975,
        17174,
        30622,
        22381,
        31148,
        23516,
        13876,
        30622,
        21372,
        5437,
        99,
        18616,
        20643,
        11,
        10907,
        22381,
        40211,
        20643,
        35748,
        32739,
        36338,
        11336,
        13594,
        29717,
        7384,
        107,
        30622,
        22381,
        12565,
        2523,
        244,
        30900,
        5437,
        99,
        18616,
        1543,
        51272
      ],
      "temperature": 0,
      "avg_logprob": -0.07672803119946552,
      "compression_ratio": 1.1163793103448276,
      "no_speech_prob": 1.9141077611806168e-11,
      "source": "whisper"
    },
    {
      "id": 7,
      "seek": 6820,
      "start": 68.2,
      "end": 76.04,
      "text": "字幕智能轉換功能可以自動把字幕文本裡的中文數字轉為阿拉伯數字,不用每條字幕手動修改。",
      "tokens": [
        50365,
        22381,
        40211,
        5094,
        118,
        8225,
        32739,
        36338,
        32311,
        8225,
        6723,
        9722,
        12572,
        16075,
        22381,
        40211,
        17174,
        8802,
        11066,
        1546,
        5975,
        17174,
        30622,
        22381,
        32739,
        6344,
        13594,
        29717,
        7384,
        107,
        30622,
        22381,
        11,
        24384,
        23664,
        31106,
        22381,
        40211,
        11389,
        12572,
        7792,
        106,
        34490,
        1543,
        50757
      ],
      "temperature": 0,
      "avg_logprob": -0.07047516552369985,
      "compression_ratio": 1.1486486486486487,
      "no_speech_prob": 2.4393308553638526e-11,
      "source": "whisper"
    },
    {
      "id": 8,
      "seek": 6820,
      "start": 76.7,
      "end": 79.66,
      "text": "轉換時會彈出轉換報告,方便核對。",
      "tokens": [
        50790,
        32739,
        36338,
        6611,
        6236,
        7391,
        230,
        7781,
        32739,
        36338,
        17803,
        16846,
        11,
        9249,
        27364,
        8987,
        116,
        2855,
        1543,
        50938
      ],
      "temperature": 0,
      "avg_logprob": -0.07047516552369985,
      "compression_ratio": 1.1486486486486487,
      "no_speech_prob": 2.4393308553638526e-11,
      "source": "whisper"
    },
    {
      "id": 9,
      "seek": 7966,
      "start": 79.66,
      "end": 88.6,
      "text": "最後,當我們把處理好的字幕導入剪輯軟件時,所有經過轉換的字幕會自動打上標記,方便快速檢查。",
      "tokens": [
        50365,
        20578,
        11,
        13118,
        5884,
        16075,
        39289,
        13876,
        20715,
        22381,
        40211,
        25647,
        14028,
        5935,
        103,
        12290,
        107,
        14454,
        253,
        20485,
        6611,
        11,
        39300,
        15145,
        8816,
        32739,
        36338,
        1546,
        22381,
        40211,
        6236,
        9722,
        12572,
        12467,
        5708,
        46299,
        16958,
        11,
        9249,
        27364,
        10251,
        31217,
        40741,
        95,
        42623,
        1543,
        50812
      ],
      "temperature": 0,
      "avg_logprob": -0.07522589189034921,
      "compression_ratio": 1.2444444444444445,
      "no_speech_prob": 2.4408752796745148e-11,
      "source": "whisper"
    },
    {
      "id": 10,
      "seek": 7966,
      "start": 92.8,
      "end": 102.8,
      "text": "針對超長字幕,我們開發了字幕智能分段功能,它會按照你設置的字符數量自動拆分,完成後彈出分割報告,方便核對。",
      "tokens": [
        51022,
        5873,
        251,
        2855,
        19869,
        15353,
        22381,
        40211,
        11,
        5884,
        8949,
        14637,
        2289,
        22381,
        40211,
        5094,
        118,
        8225,
        6627,
        28427,
        32311,
        8225,
        11,
        11284,
        6236,
        26613,
        32150,
        2166,
        39035,
        34719,
        1546,
        22381,
        5437,
        99,
        30622,
        26748,
        9722,
        12572,
        6852,
        228,
        6627,
        11,
        41509,
        5661,
        7391,
        230,
        7781,
        6627,
        5935,
        110,
        17803,
        16846,
        11,
        9249,
        27364,
        8987,
        116,
        2855,
        1543,
        51522
      ],
      "temperature": 0,
      "avg_logprob": -0.07522589189034921,
      "compression_ratio": 1.2444444444444445,
      "no_speech_prob": 2.4408752796745148e-11,
      "source": "whisper"
    },
    {
      "id": 11,
      "seek": 10966,
      "start": 109.66,
      "end": 113.66,
      "text": "以上就是本次更新的核心內容。",
      "tokens": [
        50365,
        29497,
        5620,
        8802,
        9487,
        19002,
        12560,
        1546,
        8987,
        116,
        7945,
        28472,
        25750,
        1543,
        50565
      ],
      "temperature": 0,
      "avg_logprob": -0.1605790356794993,
      "compression_ratio": 0.9629629629629629,
      "no_speech_prob": 2.7645555983335512e-11,
      "source": "whisper"
    },
    {
      "id": 12,
      "seek": 10966,
      "start": 114.3,
      "end": 117.66,
      "text": "大家遇到任何問題,都可以通過QQ群反饋。",
      "tokens": [
        50597,
        6868,
        3330,
        229,
        4511,
        26443,
        11906,
        17197,
        11,
        7182,
        6723,
        19550,
        8816,
        48,
        48,
        7248,
        97,
        22138,
        41520,
        233,
        1543,
        50765
      ],
      "temperature": 0,
      "avg_logprob": -0.1605790356794993,
      "compression_ratio": 0.9629629629629629,
      "no_speech_prob": 2.7645555983335512e-11,
      "source": "whisper"
    },
    {
      "id": 13,
      "seek": 10966,
      "start": 118.3,
      "end": 120.2,
      "text": "感謝觀看,我們下次再見。",
      "tokens": [
        50797,
        40143,
        25179,
        4200,
        11,
        5884,
        35829,
        30177,
        1543,
        50892
      ],
      "temperature": 0,
      "avg_logprob": -0.1605790356794993,
      "compression_ratio": 0.9629629629629629,
      "no_speech_prob": 2.7645555983335512e-11,
      "source": "whisper"
    }
  ],
  "transcript_selection": {
    "schema": "videonote.transcript_selection.v1",
    "status": "SELECTED",
    "requested_language": "zh",
    "duration": 120.768,
    "reference_path": null,
    "reference_kind": null,
    "reference_language": null,
    "selected_source": "whisper_segmented",
    "decision_reason": "source_unavailable_or_untrusted_whisper_passed",
    "whisper_rejected": false,
    "source_profile": {
      "segment_count": 0,
      "duration_coverage": 0.0,
      "usable_full": false,
      "usable_partial": false,
      "hallucination_region_count": 0,
      "hallucination_regions": [],
      "language": {
        "requested": "zh",
        "matched": false
      }
    },
    "whisper_profile": {
      "segment_count": 14,
      "multilingual": false,
      "first_start": 0.0,
      "final_end": 120.2,
      "duration_coverage": 0.995297,
      "maximum_gap_seconds": 6.86,
      "inverted_count": 0,
      "language": {
        "requested": "zh",
        "matched": true,
        "latin": 70,
        "han": 449,
        "kana": 0,
        "hangul": 0
      },
      "hallucination_region_count": 0,
      "hallucination_regions": [],
      "usable_full": true,
      "usable_partial": true
    },
    "comparison": {
      "window_seconds": 30.0,
      "compared_window_count": 0,
      "median_score": 0.0,
      "divergent_window_count": 0,
      "divergent_ratio": 0.0,
      "divergent_windows": []
    },
    "timestamp_alignment": {
      "applied": false,
      "estimated_offset_seconds": 0.0,
      "matched_segment_count": 0,
      "match_scores": []
    },
    "caption_timing_repair": null,
    "filled_source_profile": null,
    "whisper_duration_guard": {
      "media_duration": 120.768,
      "rejected_segment_count": 0,
      "rejected_segments": []
    }
  },
  "presegmentation_correction": {
    "applied": true,
    "correction_count": 4,
    "deterministic_fix_count": 0
  }
}