{
  "text": "我們講的是一個Worthy Agent一個小案例。今天晚上我給大家講的這個Worthy Agent,它的最大的特點就是架構不同。我們採用的是稱之為叫三明治架構。\n讲到这个三明治架构\n我首先给你解释一下\n什么叫三明治架构\n那这里要讲到\n所有市面上的\n只要是涉及到语音制农体的\n三种主流价格\nOK\n从工程的实现角度啊\n我们把所有的语音制农体\n它底层的架构分成三个\n但我这个文档我不可以呢\n一个一个给大家说啊\n这个文档我们明天\n训练结束之后的时候会发啊\n包括代码\n包括代码也会发\n好\n我们继续讲\n第一种架构\n就是我今天晚上我要讲的三明治架构\n它采用的是STT加上大语言推定模型\n加上TTS模型\n你看很明显\n看到有个这么东西\n像一个三明治一样\n把一个东西夹在中间\n把Agent加上推定模型夹在中间\n对吧\n这个就是所谓三明治架构\n三明治架构什么好处\n稍等我等会把这个讲完\n第二种是原生\n原生实时的speed to speed的架构\n原来我用的就是speed to speed的架构\n再讲一次\n这个架构就是speed to speed的架构\n那这个特点呢\n就是你得找一个\n可以实时进行语音交付的多模态模型\n这一定要是实时的\nreal time的一定是实时的\n如果你找的不是那种实时的多模态大模型\n不行\n但问题在于\n来了啊\n问题在于市面上现在很少有成熟稳定的知名的实时的多模态大模型\n包括国内包括国外\n很多人说老师不对啊\n那个那个\n那个质朴的质朴那个5.5.6 5.7\n我知道他是多模态的大模型\n他能够处理语音啊\n你要明白\n这个所说的多模特大模型\n它不单是输入是语音图片文字视频\n输出也是语音文字图片视频\n你要明白这一点\n对我们的语音这东西来说\n那输出也应该是语音\n所以实际上\n满足这个要求的\n目前我主流的那些大模型\n其实很少\n几乎没有\n目前我怎么没找到一个\n我给大家随便说几点\n这个稍微扩展一下好吧\n这个大家对模型的分类\n也一个最基本的了解\n这个所以我觉得也不是\n也不是这个扯淡\n应该是增加一下大家的知识面\n大家可以看到\n制谱里面也有\n签问里面也有\n包括国外的这个OpenAI的GPD5.6\n5.5\n很多同学说\n这不都是多模态大模型吗\n你要注意\n是的\n他们是确实实是多模态大模型\n但是这里面所说的多模态大模型\n跟我们讲的real time的多模态大模型\n多模态语音大模型还不一样\n不一样在哪里呢\n你看我们随便打开一个\n比如说假设我们很多人说\n老师你看这个5V\n质朴的5V\n或者是4.6V\n也无所谓了\n那你发现没有\n它最大的一个问题就在这\n我不知道大家看到了没有\n看到没\n它的输入没问题\n支持多模态的\n但是输出不支持多模态\n懂吗\n老师那他输出不知是多么\n他为什么就不能作为你刚刚所说的这个架构的一种模型呢\n因为语音制浓体它最重要的是实时性\n你不能说老师我语音制浓体简单\n我反正输的是语音\n最后生成的回复的也是语音就行了嘛\n没错\n确实没错\n大家讲这个是对的\n但是市面上的这个语音直农体\n跟真正意义上的\n我所说的语音直农体项目\n最大的一个问题\n就在于这个延迟性和实时性\n这个正要八经的语音直农体的延迟性\n你应该是感受不出来\n大概在零点二秒以内\n对大家听到这个词啊\n零点二秒以内\n让人类感受不出来\n那老师那那那叫你这么说\n我用质谱的这个5V\n他能感受得出来吗\n当然会\n好吧\n当然会\n如果你采用的是speed to speed的话\n他当然会\n为什么\n因为他输出的是文本\n你还需要再把文本变成什么\n变成语音\n对吧\n是不是\n好\n那还有棒连\n这个棒连\n这个棒连\n这个棒连\n这个棒连\n实时\n人生实时的一个特点\n好\n我讲完这个之后了\n当我这里没讲详细啊\n我重点是想谈两件事情\n第一件事情\n就你既然是一个智能体\n那么智能体得有循环迭代\n得有用户意图识别\n得有工具的调用\nskill的调用\n等等上下文管理\n你既然是个智能体\n你是不是这些都应该有\n没错吧\n好\n但问题在于\n如果你采用这种原生的实时的\n这种价格的方式的时候\n你发现你去找一个这样的模型\n可能能找到\n但是他对智能体的支持不是特别好\n他在智能体\n这个这种数据平台上的测试得分很低\n最近大家听到两个新闻\n第一个新闻是中国上海的一个模型\n叫来自于这个KIMI的K3\n是吧\n还有这个质朴的最新的5.2\n在国外所有这些非常知名的测试平台上\n得到的分数是非常接近于\nOpenAI的5.5对吧\n也非常接近于Antherapyc里面的\n各种4.6的这个这个Opus的4.6的这样的模型\n那大家最重要的你知道吗\n是看两个数据得分\n一个是什么数据\n一个是coding的数据\n一个是智能体的数据\n我不知道大家是否能听得懂我讲的话\n我这里面我今天没有准备好准备好那个图片\n我这个本来按道理说给大家要看那个图片的描述了\n我可能可能这个质谱里面\n这个摩塔社区里面有\n对摩塔社区里面有\n一般来说一个模型的新模型发布\n它都会有这样的一个各个数据平台的一个得分情况\n好吧\n我看一下有没有\n这个大家听懂的\nQ1啊\n就是任何一个牛逼的模型\n之所以在全国来说\n是比较高的一个\n比较比较高知名度的\n他有两个分数很重要\n一个是这个分数\n大家看\n这个分数\n好吧\n还有一个是关于coding的那个分数\n但最正的是这个\n大家为什么要重点看这个指标吗\n看这个基准测试的数据\n这个基准测试的数据集的测试结果\n测试结果的分数吗\n为什么大家看这个\n大家知道不\n就不用我再说了吧\n好吧\n大家为什么要看这个指标\n其他的不看\n或者其他的很少看\n不是完全不看啊\n其实我们跟国外的模型比较\n重要的就比较两个最重要的指标\n一个是这个智能体的分数\n智能体的能力\n智能体的准确性的能力\n还有一个是一个coding能力\n就这两个\n你发现,你像DeepSync V4,Flash 0731,这个正式版,你发现他是不是非常接近于Opus Anthropic的Opus 4.8,对吧?\n没错吧\n所以我们国外也好\n国内也好\n大家都知道\n这个DeepSync V4的正式版本\n好像跟这个Opus 4.8\n是相接近\n其实不单是DeepSync\n还有那个刚刚我说的Kimin的Key3\n以及智朴的最新发布的那个模型\n是吧\n对吧\n那我那老师你讲这话什么意思呢\n是这样子的\n就是你的智能\n你要你要想作为智能体的核心推敌模型\n那么你一定要找\n你一定要去找\n这个分数比较高的那些模型\n懂意思吧\n现在我明白了吗\n那么这个分数比较高的那些模型\n都是什么\n这个\n这个\n这个\n这个\n是吧\n当然还有这个OpenAI的\n这个GPT5.5 5.6\n那个\n那个\n那个\n他没列在这里头\n对吧\n哦\n明白了\n老师少说大概你懂意思了\n就是说\n我如果要开发一个流逼的智能体\n那我智能体里面的核心推理模型\n和核心的意图识别模型\n得是很强的\n而且怎么证明它强\n就是看这个模型\n在这个数据级上的基准测试的得分来判断的\n是吧\n是的\n没错是这个意思\n好\n接下来我再讲另外一句话\n好\n那有一些\n有一些全模态的大模型\n他虽然也能够支持语音的输入\n同时也支持语音的输出\n但问题在于\n他在这个数据级的基准测试的得分\n是非常非常的低的\n举个例子\n比如说像全模态大模型\n有一个\n就是我们经常讲的\n签文里面有一个\n稍等啊\n搜索一下\n看是能搜到啊\n这个搜索框去了\n这个\n好吧\n这就是千万里面一个非常典型的全模态大模型\n什么叫全模态大模型呢\n就是\n这个网页打不开啊\n就是他输入的可以是语音\n然后他可以思考\n思考完之后\n它输出的结果\n也可以是语音\n也可以是文字\n语音文字都支持\n那这种模型\n是不是非常适合\n我们做这种\n这个语音的智能体\n是的\n非常适合的\n但问题在于\n这样的模型\n它在智能体\n这个数据级上的评测得分\n是远远低于\n大家看到的\n这些主流的推定模型\n现在明白意思吧\n所以\n在实际的企业开发过程中\n如果要做一个语音模型\n你又想要保证实时性\n你又想保证我这个智能体核心的推理模型和意图识别模型\n是很好的模型\n是很牛的模型\n那这两个要求都要满足\n说实话\n很难有完美的方案\n所以为了同时满足这两个要求\n我们今天重新\n根据当了我前两天也是看到网上论文\n根据网上论文\n我构建了一个新的方案\n叫三明治方案\n那三明治方案它的区别是\n它的特点是什么呢\n就是重点放在智能体这一块\n智能体的核心推理模型\n我还是使用这些像\nDeepSync V4的正式版\n或者Opus的4.8\n或者是OpenAI的5.6\n5.5这样的模型\n或者是质朴的\n对吧最新的模型\n好然后呢\n但这两个\n他在智能体这方面\n在智能体这个\n这个基准测试数据上表现分数非常高\n对吧还不错非常好\n可是问题是他对语音的输入和输出\n不是特别支持\n对吧\n所以就有一个这么一个矛盾点\n那怎么办呢\n我在它的前面\n加上一个STD模型\n我在它的后面\n加上一个TTS模型\n把它夹到中间\n就称之为这样\n三米之架构\n现在我应该讲清楚了",
  "language": "zh",
  "duration": 756.41325,
  "segments": [
    {
      "start": 0.0,
      "end": 17.0,
      "text": "我們講的是一個Worthy Agent一個小案例。今天晚上我給大家講的這個Worthy Agent,它的最大的特點就是架構不同。我們採用的是稱之為叫三明治架構。",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.3034486603318599,
      "no_speech_prob": 1.1696877841305131e-11,
      "compression_ratio": 1.1242603550295858,
      "source": "whisper"
    },
    {
      "start": 17.0,
      "end": 18.56,
      "text": "讲到这个三明治架构",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.16405868530273438,
      "no_speech_prob": 1.9598952669119818e-11,
      "compression_ratio": 1.3900709219858156,
      "source": "whisper"
    },
    {
      "start": 18.56,
      "end": 19.9,
      "text": "我首先给你解释一下",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.16405868530273438,
      "no_speech_prob": 1.9598952669119818e-11,
      "compression_ratio": 1.3900709219858156,
      "source": "whisper"
    },
    {
      "start": 19.9,
      "end": 21.76,
      "text": "什么叫三明治架构",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.16405868530273438,
      "no_speech_prob": 1.9598952669119818e-11,
      "compression_ratio": 1.3900709219858156,
      "source": "whisper"
    },
    {
      "start": 21.76,
      "end": 23.34,
      "text": "那这里要讲到",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.16405868530273438,
      "no_speech_prob": 1.9598952669119818e-11,
      "compression_ratio": 1.3900709219858156,
      "source": "whisper"
    },
    {
      "start": 23.34,
      "end": 25.12,
      "text": "所有市面上的",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.16405868530273438,
      "no_speech_prob": 1.9598952669119818e-11,
      "compression_ratio": 1.3900709219858156,
      "source": "whisper"
    },
    {
      "start": 25.12,
      "end": 28.36,
      "text": "只要是涉及到语音制农体的",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.16405868530273438,
      "no_speech_prob": 1.9598952669119818e-11,
      "compression_ratio": 1.3900709219858156,
      "source": "whisper"
    },
    {
      "start": 28.36,
      "end": 29.96,
      "text": "三种主流价格",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.16405868530273438,
      "no_speech_prob": 1.9598952669119818e-11,
      "compression_ratio": 1.3900709219858156,
      "source": "whisper"
    },
    {
      "start": 29.96,
      "end": 31.26,
      "text": "OK",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.16405868530273438,
      "no_speech_prob": 1.9598952669119818e-11,
      "compression_ratio": 1.3900709219858156,
      "source": "whisper"
    },
    {
      "start": 31.26,
      "end": 33.5,
      "text": "从工程的实现角度啊",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.16405868530273438,
      "no_speech_prob": 1.9598952669119818e-11,
      "compression_ratio": 1.3900709219858156,
      "source": "whisper"
    },
    {
      "start": 33.5,
      "end": 35.68,
      "text": "我们把所有的语音制农体",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.16405868530273438,
      "no_speech_prob": 1.9598952669119818e-11,
      "compression_ratio": 1.3900709219858156,
      "source": "whisper"
    },
    {
      "start": 35.68,
      "end": 38.68,
      "text": "它底层的架构分成三个",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.16405868530273438,
      "no_speech_prob": 1.9598952669119818e-11,
      "compression_ratio": 1.3900709219858156,
      "source": "whisper"
    },
    {
      "start": 38.68,
      "end": 40.54,
      "text": "但我这个文档我不可以呢",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.16405868530273438,
      "no_speech_prob": 1.9598952669119818e-11,
      "compression_ratio": 1.3900709219858156,
      "source": "whisper"
    },
    {
      "start": 40.54,
      "end": 41.94,
      "text": "一个一个给大家说啊",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.16405868530273438,
      "no_speech_prob": 1.9598952669119818e-11,
      "compression_ratio": 1.3900709219858156,
      "source": "whisper"
    },
    {
      "start": 41.94,
      "end": 43.32,
      "text": "这个文档我们明天",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.16405868530273438,
      "no_speech_prob": 1.9598952669119818e-11,
      "compression_ratio": 1.3900709219858156,
      "source": "whisper"
    },
    {
      "start": 43.32,
      "end": 45.88,
      "text": "训练结束之后的时候会发啊",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.16405868530273438,
      "no_speech_prob": 1.9598952669119818e-11,
      "compression_ratio": 1.3900709219858156,
      "source": "whisper"
    },
    {
      "start": 45.88,
      "end": 46.4,
      "text": "包括代码",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.16405868530273438,
      "no_speech_prob": 1.9598952669119818e-11,
      "compression_ratio": 1.3900709219858156,
      "source": "whisper"
    },
    {
      "start": 46.4,
      "end": 47.22,
      "text": "包括代码也会发",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.15370556403850688,
      "no_speech_prob": 2.6877725739504754e-11,
      "compression_ratio": 1.4532374100719425,
      "source": "whisper"
    },
    {
      "start": 47.22,
      "end": 48.04,
      "text": "好",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.15370556403850688,
      "no_speech_prob": 2.6877725739504754e-11,
      "compression_ratio": 1.4532374100719425,
      "source": "whisper"
    },
    {
      "start": 48.04,
      "end": 49.3,
      "text": "我们继续讲",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.15370556403850688,
      "no_speech_prob": 2.6877725739504754e-11,
      "compression_ratio": 1.4532374100719425,
      "source": "whisper"
    },
    {
      "start": 49.3,
      "end": 50.62,
      "text": "第一种架构",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.15370556403850688,
      "no_speech_prob": 2.6877725739504754e-11,
      "compression_ratio": 1.4532374100719425,
      "source": "whisper"
    },
    {
      "start": 50.62,
      "end": 52.74,
      "text": "就是我今天晚上我要讲的三明治架构",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.15370556403850688,
      "no_speech_prob": 2.6877725739504754e-11,
      "compression_ratio": 1.4532374100719425,
      "source": "whisper"
    },
    {
      "start": 52.74,
      "end": 56.52,
      "text": "它采用的是STT加上大语言推定模型",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.15370556403850688,
      "no_speech_prob": 2.6877725739504754e-11,
      "compression_ratio": 1.4532374100719425,
      "source": "whisper"
    },
    {
      "start": 56.52,
      "end": 58.26,
      "text": "加上TTS模型",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.15370556403850688,
      "no_speech_prob": 2.6877725739504754e-11,
      "compression_ratio": 1.4532374100719425,
      "source": "whisper"
    },
    {
      "start": 58.26,
      "end": 59.78,
      "text": "你看很明显",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.15370556403850688,
      "no_speech_prob": 2.6877725739504754e-11,
      "compression_ratio": 1.4532374100719425,
      "source": "whisper"
    },
    {
      "start": 59.78,
      "end": 60.82,
      "text": "看到有个这么东西",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.15370556403850688,
      "no_speech_prob": 2.6877725739504754e-11,
      "compression_ratio": 1.4532374100719425,
      "source": "whisper"
    },
    {
      "start": 60.82,
      "end": 62.2,
      "text": "像一个三明治一样",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.15370556403850688,
      "no_speech_prob": 2.6877725739504754e-11,
      "compression_ratio": 1.4532374100719425,
      "source": "whisper"
    },
    {
      "start": 62.2,
      "end": 64.04,
      "text": "把一个东西夹在中间",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.15370556403850688,
      "no_speech_prob": 2.6877725739504754e-11,
      "compression_ratio": 1.4532374100719425,
      "source": "whisper"
    },
    {
      "start": 64.04,
      "end": 67.5,
      "text": "把Agent加上推定模型夹在中间",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.15370556403850688,
      "no_speech_prob": 2.6877725739504754e-11,
      "compression_ratio": 1.4532374100719425,
      "source": "whisper"
    },
    {
      "start": 67.5,
      "end": 67.9,
      "text": "对吧",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.15370556403850688,
      "no_speech_prob": 2.6877725739504754e-11,
      "compression_ratio": 1.4532374100719425,
      "source": "whisper"
    },
    {
      "start": 67.9,
      "end": 70.26,
      "text": "这个就是所谓三明治架构",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.15370556403850688,
      "no_speech_prob": 2.6877725739504754e-11,
      "compression_ratio": 1.4532374100719425,
      "source": "whisper"
    },
    {
      "start": 70.26,
      "end": 72.3,
      "text": "三明治架构什么好处",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.15370556403850688,
      "no_speech_prob": 2.6877725739504754e-11,
      "compression_ratio": 1.4532374100719425,
      "source": "whisper"
    },
    {
      "start": 72.3,
      "end": 73.88,
      "text": "稍等我等会把这个讲完",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.15370556403850688,
      "no_speech_prob": 2.6877725739504754e-11,
      "compression_ratio": 1.4532374100719425,
      "source": "whisper"
    },
    {
      "start": 73.88,
      "end": 75.92,
      "text": "第二种是原生",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.15370556403850688,
      "no_speech_prob": 2.6877725739504754e-11,
      "compression_ratio": 1.4532374100719425,
      "source": "whisper"
    },
    {
      "start": 75.92,
      "end": 79.56,
      "text": "原生实时的speed to speed的架构",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1596087147207821,
      "no_speech_prob": 1.836425629619942e-11,
      "compression_ratio": 1.4937759336099585,
      "source": "whisper"
    },
    {
      "start": 79.56,
      "end": 82.0,
      "text": "原来我用的就是speed to speed的架构",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1596087147207821,
      "no_speech_prob": 1.836425629619942e-11,
      "compression_ratio": 1.4937759336099585,
      "source": "whisper"
    },
    {
      "start": 82.0,
      "end": 82.54,
      "text": "再讲一次",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1596087147207821,
      "no_speech_prob": 1.836425629619942e-11,
      "compression_ratio": 1.4937759336099585,
      "source": "whisper"
    },
    {
      "start": 82.54,
      "end": 84.68,
      "text": "这个架构就是speed to speed的架构",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1596087147207821,
      "no_speech_prob": 1.836425629619942e-11,
      "compression_ratio": 1.4937759336099585,
      "source": "whisper"
    },
    {
      "start": 84.68,
      "end": 86.26,
      "text": "那这个特点呢",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1596087147207821,
      "no_speech_prob": 1.836425629619942e-11,
      "compression_ratio": 1.4937759336099585,
      "source": "whisper"
    },
    {
      "start": 86.26,
      "end": 87.5,
      "text": "就是你得找一个",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1596087147207821,
      "no_speech_prob": 1.836425629619942e-11,
      "compression_ratio": 1.4937759336099585,
      "source": "whisper"
    },
    {
      "start": 87.5,
      "end": 91.86,
      "text": "可以实时进行语音交付的多模态模型",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1596087147207821,
      "no_speech_prob": 1.836425629619942e-11,
      "compression_ratio": 1.4937759336099585,
      "source": "whisper"
    },
    {
      "start": 91.86,
      "end": 94.48,
      "text": "这一定要是实时的",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1596087147207821,
      "no_speech_prob": 1.836425629619942e-11,
      "compression_ratio": 1.4937759336099585,
      "source": "whisper"
    },
    {
      "start": 94.48,
      "end": 97.64,
      "text": "real time的一定是实时的",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1596087147207821,
      "no_speech_prob": 1.836425629619942e-11,
      "compression_ratio": 1.4937759336099585,
      "source": "whisper"
    },
    {
      "start": 97.64,
      "end": 101.26,
      "text": "如果你找的不是那种实时的多模态大模型",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1596087147207821,
      "no_speech_prob": 1.836425629619942e-11,
      "compression_ratio": 1.4937759336099585,
      "source": "whisper"
    },
    {
      "start": 101.26,
      "end": 101.6,
      "text": "不行",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1596087147207821,
      "no_speech_prob": 1.836425629619942e-11,
      "compression_ratio": 1.4937759336099585,
      "source": "whisper"
    },
    {
      "start": 101.6,
      "end": 103.82,
      "text": "但问题在于",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1596087147207821,
      "no_speech_prob": 1.836425629619942e-11,
      "compression_ratio": 1.4937759336099585,
      "source": "whisper"
    },
    {
      "start": 103.82,
      "end": 104.74,
      "text": "来了啊",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1596087147207821,
      "no_speech_prob": 1.836425629619942e-11,
      "compression_ratio": 1.4937759336099585,
      "source": "whisper"
    },
    {
      "start": 104.74,
      "end": 113.82,
      "text": "问题在于市面上现在很少有成熟稳定的知名的实时的多模态大模型",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.18331880852727606,
      "no_speech_prob": 1.253278297114191e-11,
      "compression_ratio": 1.2746113989637307,
      "source": "whisper"
    },
    {
      "start": 113.82,
      "end": 115.88,
      "text": "包括国内包括国外",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.18331880852727606,
      "no_speech_prob": 1.253278297114191e-11,
      "compression_ratio": 1.2746113989637307,
      "source": "whisper"
    },
    {
      "start": 115.88,
      "end": 118.98,
      "text": "很多人说老师不对啊",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.18331880852727606,
      "no_speech_prob": 1.253278297114191e-11,
      "compression_ratio": 1.2746113989637307,
      "source": "whisper"
    },
    {
      "start": 118.98,
      "end": 120.1,
      "text": "那个那个",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.18331880852727606,
      "no_speech_prob": 1.253278297114191e-11,
      "compression_ratio": 1.2746113989637307,
      "source": "whisper"
    },
    {
      "start": 120.1,
      "end": 126.34,
      "text": "那个质朴的质朴那个5.5.6 5.7",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.18331880852727606,
      "no_speech_prob": 1.253278297114191e-11,
      "compression_ratio": 1.2746113989637307,
      "source": "whisper"
    },
    {
      "start": 126.34,
      "end": 129.32,
      "text": "我知道他是多模态的大模型",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.18331880852727606,
      "no_speech_prob": 1.253278297114191e-11,
      "compression_ratio": 1.2746113989637307,
      "source": "whisper"
    },
    {
      "start": 129.32,
      "end": 131.98,
      "text": "他能够处理语音啊",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.18331880852727606,
      "no_speech_prob": 1.253278297114191e-11,
      "compression_ratio": 1.2746113989637307,
      "source": "whisper"
    },
    {
      "start": 131.98,
      "end": 132.64,
      "text": "你要明白",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.15738445723137887,
      "no_speech_prob": 2.0169553527349393e-11,
      "compression_ratio": 1.457142857142857,
      "source": "whisper"
    },
    {
      "start": 132.64,
      "end": 134.7,
      "text": "这个所说的多模特大模型",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.15738445723137887,
      "no_speech_prob": 2.0169553527349393e-11,
      "compression_ratio": 1.457142857142857,
      "source": "whisper"
    },
    {
      "start": 134.7,
      "end": 138.98,
      "text": "它不单是输入是语音图片文字视频",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.15738445723137887,
      "no_speech_prob": 2.0169553527349393e-11,
      "compression_ratio": 1.457142857142857,
      "source": "whisper"
    },
    {
      "start": 138.98,
      "end": 142.16,
      "text": "输出也是语音文字图片视频",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.15738445723137887,
      "no_speech_prob": 2.0169553527349393e-11,
      "compression_ratio": 1.457142857142857,
      "source": "whisper"
    },
    {
      "start": 142.16,
      "end": 143.72,
      "text": "你要明白这一点",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.15738445723137887,
      "no_speech_prob": 2.0169553527349393e-11,
      "compression_ratio": 1.457142857142857,
      "source": "whisper"
    },
    {
      "start": 143.72,
      "end": 146.88,
      "text": "对我们的语音这东西来说",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.15738445723137887,
      "no_speech_prob": 2.0169553527349393e-11,
      "compression_ratio": 1.457142857142857,
      "source": "whisper"
    },
    {
      "start": 146.88,
      "end": 149.4,
      "text": "那输出也应该是语音",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.15738445723137887,
      "no_speech_prob": 2.0169553527349393e-11,
      "compression_ratio": 1.457142857142857,
      "source": "whisper"
    },
    {
      "start": 149.4,
      "end": 151.58,
      "text": "所以实际上",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.15738445723137887,
      "no_speech_prob": 2.0169553527349393e-11,
      "compression_ratio": 1.457142857142857,
      "source": "whisper"
    },
    {
      "start": 151.58,
      "end": 153.66,
      "text": "满足这个要求的",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.15738445723137887,
      "no_speech_prob": 2.0169553527349393e-11,
      "compression_ratio": 1.457142857142857,
      "source": "whisper"
    },
    {
      "start": 153.66,
      "end": 156.22,
      "text": "目前我主流的那些大模型",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.15738445723137887,
      "no_speech_prob": 2.0169553527349393e-11,
      "compression_ratio": 1.457142857142857,
      "source": "whisper"
    },
    {
      "start": 156.22,
      "end": 156.86,
      "text": "其实很少",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.15738445723137887,
      "no_speech_prob": 2.0169553527349393e-11,
      "compression_ratio": 1.457142857142857,
      "source": "whisper"
    },
    {
      "start": 156.86,
      "end": 157.62,
      "text": "几乎没有",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.15738445723137887,
      "no_speech_prob": 2.0169553527349393e-11,
      "compression_ratio": 1.457142857142857,
      "source": "whisper"
    },
    {
      "start": 157.62,
      "end": 159.84,
      "text": "目前我怎么没找到一个",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.15738445723137887,
      "no_speech_prob": 2.0169553527349393e-11,
      "compression_ratio": 1.457142857142857,
      "source": "whisper"
    },
    {
      "start": 159.84,
      "end": 161.56,
      "text": "我给大家随便说几点",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.15738445723137887,
      "no_speech_prob": 2.0169553527349393e-11,
      "compression_ratio": 1.457142857142857,
      "source": "whisper"
    },
    {
      "start": 161.56,
      "end": 163.14,
      "text": "这个稍微扩展一下好吧",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1393848949008518,
      "no_speech_prob": 2.315848728062786e-11,
      "compression_ratio": 1.4482758620689655,
      "source": "whisper"
    },
    {
      "start": 163.14,
      "end": 165.78,
      "text": "这个大家对模型的分类",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1393848949008518,
      "no_speech_prob": 2.315848728062786e-11,
      "compression_ratio": 1.4482758620689655,
      "source": "whisper"
    },
    {
      "start": 165.78,
      "end": 166.84,
      "text": "也一个最基本的了解",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1393848949008518,
      "no_speech_prob": 2.315848728062786e-11,
      "compression_ratio": 1.4482758620689655,
      "source": "whisper"
    },
    {
      "start": 166.84,
      "end": 168.4,
      "text": "这个所以我觉得也不是",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1393848949008518,
      "no_speech_prob": 2.315848728062786e-11,
      "compression_ratio": 1.4482758620689655,
      "source": "whisper"
    },
    {
      "start": 168.4,
      "end": 170.86,
      "text": "也不是这个扯淡",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1393848949008518,
      "no_speech_prob": 2.315848728062786e-11,
      "compression_ratio": 1.4482758620689655,
      "source": "whisper"
    },
    {
      "start": 170.86,
      "end": 173.12,
      "text": "应该是增加一下大家的知识面",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1393848949008518,
      "no_speech_prob": 2.315848728062786e-11,
      "compression_ratio": 1.4482758620689655,
      "source": "whisper"
    },
    {
      "start": 173.12,
      "end": 174.42,
      "text": "大家可以看到",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1393848949008518,
      "no_speech_prob": 2.315848728062786e-11,
      "compression_ratio": 1.4482758620689655,
      "source": "whisper"
    },
    {
      "start": 174.42,
      "end": 175.76,
      "text": "制谱里面也有",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1393848949008518,
      "no_speech_prob": 2.315848728062786e-11,
      "compression_ratio": 1.4482758620689655,
      "source": "whisper"
    },
    {
      "start": 175.76,
      "end": 176.68,
      "text": "签问里面也有",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1393848949008518,
      "no_speech_prob": 2.315848728062786e-11,
      "compression_ratio": 1.4482758620689655,
      "source": "whisper"
    },
    {
      "start": 176.68,
      "end": 179.42,
      "text": "包括国外的这个OpenAI的GPD5.6",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1393848949008518,
      "no_speech_prob": 2.315848728062786e-11,
      "compression_ratio": 1.4482758620689655,
      "source": "whisper"
    },
    {
      "start": 179.42,
      "end": 180.12,
      "text": "5.5",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1393848949008518,
      "no_speech_prob": 2.315848728062786e-11,
      "compression_ratio": 1.4482758620689655,
      "source": "whisper"
    },
    {
      "start": 180.12,
      "end": 181.1,
      "text": "很多同学说",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1393848949008518,
      "no_speech_prob": 2.315848728062786e-11,
      "compression_ratio": 1.4482758620689655,
      "source": "whisper"
    },
    {
      "start": 181.1,
      "end": 182.84,
      "text": "这不都是多模态大模型吗",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1393848949008518,
      "no_speech_prob": 2.315848728062786e-11,
      "compression_ratio": 1.4482758620689655,
      "source": "whisper"
    },
    {
      "start": 182.84,
      "end": 183.52,
      "text": "你要注意",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1393848949008518,
      "no_speech_prob": 2.315848728062786e-11,
      "compression_ratio": 1.4482758620689655,
      "source": "whisper"
    },
    {
      "start": 183.52,
      "end": 184.04,
      "text": "是的",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1393848949008518,
      "no_speech_prob": 2.315848728062786e-11,
      "compression_ratio": 1.4482758620689655,
      "source": "whisper"
    },
    {
      "start": 184.04,
      "end": 186.04,
      "text": "他们是确实实是多模态大模型",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1393848949008518,
      "no_speech_prob": 2.315848728062786e-11,
      "compression_ratio": 1.4482758620689655,
      "source": "whisper"
    },
    {
      "start": 186.04,
      "end": 188.06,
      "text": "但是这里面所说的多模态大模型",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1393848949008518,
      "no_speech_prob": 2.315848728062786e-11,
      "compression_ratio": 1.4482758620689655,
      "source": "whisper"
    },
    {
      "start": 188.06,
      "end": 190.68,
      "text": "跟我们讲的real time的多模态大模型",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1393848949008518,
      "no_speech_prob": 2.315848728062786e-11,
      "compression_ratio": 1.4482758620689655,
      "source": "whisper"
    },
    {
      "start": 190.68,
      "end": 192.9,
      "text": "多模态语音大模型还不一样",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.17897313706418302,
      "no_speech_prob": 1.94090767763333e-11,
      "compression_ratio": 1.350194552529183,
      "source": "whisper"
    },
    {
      "start": 192.9,
      "end": 194.2,
      "text": "不一样在哪里呢",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.17897313706418302,
      "no_speech_prob": 1.94090767763333e-11,
      "compression_ratio": 1.350194552529183,
      "source": "whisper"
    },
    {
      "start": 194.2,
      "end": 195.3,
      "text": "你看我们随便打开一个",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.17897313706418302,
      "no_speech_prob": 1.94090767763333e-11,
      "compression_ratio": 1.350194552529183,
      "source": "whisper"
    },
    {
      "start": 195.3,
      "end": 197.74,
      "text": "比如说假设我们很多人说",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.17897313706418302,
      "no_speech_prob": 1.94090767763333e-11,
      "compression_ratio": 1.350194552529183,
      "source": "whisper"
    },
    {
      "start": 197.74,
      "end": 199.16,
      "text": "老师你看这个5V",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.17897313706418302,
      "no_speech_prob": 1.94090767763333e-11,
      "compression_ratio": 1.350194552529183,
      "source": "whisper"
    },
    {
      "start": 199.16,
      "end": 201.04,
      "text": "质朴的5V",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.17897313706418302,
      "no_speech_prob": 1.94090767763333e-11,
      "compression_ratio": 1.350194552529183,
      "source": "whisper"
    },
    {
      "start": 201.04,
      "end": 203.14,
      "text": "或者是4.6V",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.17897313706418302,
      "no_speech_prob": 1.94090767763333e-11,
      "compression_ratio": 1.350194552529183,
      "source": "whisper"
    },
    {
      "start": 203.14,
      "end": 203.88,
      "text": "也无所谓了",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.17897313706418302,
      "no_speech_prob": 1.94090767763333e-11,
      "compression_ratio": 1.350194552529183,
      "source": "whisper"
    },
    {
      "start": 203.88,
      "end": 205.14,
      "text": "那你发现没有",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.17897313706418302,
      "no_speech_prob": 1.94090767763333e-11,
      "compression_ratio": 1.350194552529183,
      "source": "whisper"
    },
    {
      "start": 205.14,
      "end": 206.94,
      "text": "它最大的一个问题就在这",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.17897313706418302,
      "no_speech_prob": 1.94090767763333e-11,
      "compression_ratio": 1.350194552529183,
      "source": "whisper"
    },
    {
      "start": 206.94,
      "end": 211.2,
      "text": "我不知道大家看到了没有",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.17897313706418302,
      "no_speech_prob": 1.94090767763333e-11,
      "compression_ratio": 1.350194552529183,
      "source": "whisper"
    },
    {
      "start": 211.2,
      "end": 213.48,
      "text": "看到没",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.17897313706418302,
      "no_speech_prob": 1.94090767763333e-11,
      "compression_ratio": 1.350194552529183,
      "source": "whisper"
    },
    {
      "start": 213.48,
      "end": 215.38,
      "text": "它的输入没问题",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.17897313706418302,
      "no_speech_prob": 1.94090767763333e-11,
      "compression_ratio": 1.350194552529183,
      "source": "whisper"
    },
    {
      "start": 215.38,
      "end": 216.32,
      "text": "支持多模态的",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.17897313706418302,
      "no_speech_prob": 1.94090767763333e-11,
      "compression_ratio": 1.350194552529183,
      "source": "whisper"
    },
    {
      "start": 216.32,
      "end": 218.1,
      "text": "但是输出不支持多模态",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.17897313706418302,
      "no_speech_prob": 1.94090767763333e-11,
      "compression_ratio": 1.350194552529183,
      "source": "whisper"
    },
    {
      "start": 218.1,
      "end": 219.32,
      "text": "懂吗",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.17897313706418302,
      "no_speech_prob": 1.94090767763333e-11,
      "compression_ratio": 1.350194552529183,
      "source": "whisper"
    },
    {
      "start": 219.32,
      "end": 222.64,
      "text": "老师那他输出不知是多么",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1577636964859501,
      "no_speech_prob": 1.306091866604131e-11,
      "compression_ratio": 1.3506493506493507,
      "source": "whisper"
    },
    {
      "start": 222.64,
      "end": 227.64,
      "text": "他为什么就不能作为你刚刚所说的这个架构的一种模型呢",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1577636964859501,
      "no_speech_prob": 1.306091866604131e-11,
      "compression_ratio": 1.3506493506493507,
      "source": "whisper"
    },
    {
      "start": 227.64,
      "end": 233.08,
      "text": "因为语音制浓体它最重要的是实时性",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1577636964859501,
      "no_speech_prob": 1.306091866604131e-11,
      "compression_ratio": 1.3506493506493507,
      "source": "whisper"
    },
    {
      "start": 233.08,
      "end": 237.66,
      "text": "你不能说老师我语音制浓体简单",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1577636964859501,
      "no_speech_prob": 1.306091866604131e-11,
      "compression_ratio": 1.3506493506493507,
      "source": "whisper"
    },
    {
      "start": 237.66,
      "end": 239.72,
      "text": "我反正输的是语音",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1577636964859501,
      "no_speech_prob": 1.306091866604131e-11,
      "compression_ratio": 1.3506493506493507,
      "source": "whisper"
    },
    {
      "start": 239.72,
      "end": 242.78,
      "text": "最后生成的回复的也是语音就行了嘛",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1577636964859501,
      "no_speech_prob": 1.306091866604131e-11,
      "compression_ratio": 1.3506493506493507,
      "source": "whisper"
    },
    {
      "start": 242.78,
      "end": 243.86,
      "text": "没错",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1577636964859501,
      "no_speech_prob": 1.306091866604131e-11,
      "compression_ratio": 1.3506493506493507,
      "source": "whisper"
    },
    {
      "start": 243.86,
      "end": 245.74,
      "text": "确实没错",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1577636964859501,
      "no_speech_prob": 1.306091866604131e-11,
      "compression_ratio": 1.3506493506493507,
      "source": "whisper"
    },
    {
      "start": 245.74,
      "end": 247.66,
      "text": "大家讲这个是对的",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1577636964859501,
      "no_speech_prob": 1.306091866604131e-11,
      "compression_ratio": 1.3506493506493507,
      "source": "whisper"
    },
    {
      "start": 247.66,
      "end": 250.7,
      "text": "但是市面上的这个语音直农体",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.13184958736912733,
      "no_speech_prob": 1.4118016651576415e-11,
      "compression_ratio": 1.4811715481171548,
      "source": "whisper"
    },
    {
      "start": 250.7,
      "end": 252.58,
      "text": "跟真正意义上的",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.13184958736912733,
      "no_speech_prob": 1.4118016651576415e-11,
      "compression_ratio": 1.4811715481171548,
      "source": "whisper"
    },
    {
      "start": 252.58,
      "end": 254.1,
      "text": "我所说的语音直农体项目",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.13184958736912733,
      "no_speech_prob": 1.4118016651576415e-11,
      "compression_ratio": 1.4811715481171548,
      "source": "whisper"
    },
    {
      "start": 254.1,
      "end": 255.24,
      "text": "最大的一个问题",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.13184958736912733,
      "no_speech_prob": 1.4118016651576415e-11,
      "compression_ratio": 1.4811715481171548,
      "source": "whisper"
    },
    {
      "start": 255.24,
      "end": 258.12,
      "text": "就在于这个延迟性和实时性",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.13184958736912733,
      "no_speech_prob": 1.4118016651576415e-11,
      "compression_ratio": 1.4811715481171548,
      "source": "whisper"
    },
    {
      "start": 258.12,
      "end": 262.24,
      "text": "这个正要八经的语音直农体的延迟性",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.13184958736912733,
      "no_speech_prob": 1.4118016651576415e-11,
      "compression_ratio": 1.4811715481171548,
      "source": "whisper"
    },
    {
      "start": 262.24,
      "end": 263.52,
      "text": "你应该是感受不出来",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.13184958736912733,
      "no_speech_prob": 1.4118016651576415e-11,
      "compression_ratio": 1.4811715481171548,
      "source": "whisper"
    },
    {
      "start": 263.52,
      "end": 267.2,
      "text": "大概在零点二秒以内",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.13184958736912733,
      "no_speech_prob": 1.4118016651576415e-11,
      "compression_ratio": 1.4811715481171548,
      "source": "whisper"
    },
    {
      "start": 267.2,
      "end": 269.24,
      "text": "对大家听到这个词啊",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.13184958736912733,
      "no_speech_prob": 1.4118016651576415e-11,
      "compression_ratio": 1.4811715481171548,
      "source": "whisper"
    },
    {
      "start": 269.24,
      "end": 271.66,
      "text": "零点二秒以内",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.13184958736912733,
      "no_speech_prob": 1.4118016651576415e-11,
      "compression_ratio": 1.4811715481171548,
      "source": "whisper"
    },
    {
      "start": 271.66,
      "end": 273.66,
      "text": "让人类感受不出来",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.13184958736912733,
      "no_speech_prob": 1.4118016651576415e-11,
      "compression_ratio": 1.4811715481171548,
      "source": "whisper"
    },
    {
      "start": 273.66,
      "end": 276.9,
      "text": "那老师那那那叫你这么说",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.13184958736912733,
      "no_speech_prob": 1.4118016651576415e-11,
      "compression_ratio": 1.4811715481171548,
      "source": "whisper"
    },
    {
      "start": 276.9,
      "end": 278.5,
      "text": "我用质谱的这个5V",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.332143119211947,
      "no_speech_prob": 2.2597562709947283e-11,
      "compression_ratio": 1.946078431372549,
      "source": "whisper"
    },
    {
      "start": 278.5,
      "end": 279.74,
      "text": "他能感受得出来吗",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.332143119211947,
      "no_speech_prob": 2.2597562709947283e-11,
      "compression_ratio": 1.946078431372549,
      "source": "whisper"
    },
    {
      "start": 279.74,
      "end": 280.66,
      "text": "当然会",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.332143119211947,
      "no_speech_prob": 2.2597562709947283e-11,
      "compression_ratio": 1.946078431372549,
      "source": "whisper"
    },
    {
      "start": 280.66,
      "end": 282.42,
      "text": "好吧",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.332143119211947,
      "no_speech_prob": 2.2597562709947283e-11,
      "compression_ratio": 1.946078431372549,
      "source": "whisper"
    },
    {
      "start": 282.42,
      "end": 283.2,
      "text": "当然会",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.332143119211947,
      "no_speech_prob": 2.2597562709947283e-11,
      "compression_ratio": 1.946078431372549,
      "source": "whisper"
    },
    {
      "start": 283.2,
      "end": 285.08,
      "text": "如果你采用的是speed to speed的话",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.332143119211947,
      "no_speech_prob": 2.2597562709947283e-11,
      "compression_ratio": 1.946078431372549,
      "source": "whisper"
    },
    {
      "start": 285.08,
      "end": 285.8,
      "text": "他当然会",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.332143119211947,
      "no_speech_prob": 2.2597562709947283e-11,
      "compression_ratio": 1.946078431372549,
      "source": "whisper"
    },
    {
      "start": 285.8,
      "end": 286.08,
      "text": "为什么",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.332143119211947,
      "no_speech_prob": 2.2597562709947283e-11,
      "compression_ratio": 1.946078431372549,
      "source": "whisper"
    },
    {
      "start": 286.08,
      "end": 287.98,
      "text": "因为他输出的是文本",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.332143119211947,
      "no_speech_prob": 2.2597562709947283e-11,
      "compression_ratio": 1.946078431372549,
      "source": "whisper"
    },
    {
      "start": 287.98,
      "end": 290.04,
      "text": "你还需要再把文本变成什么",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.332143119211947,
      "no_speech_prob": 2.2597562709947283e-11,
      "compression_ratio": 1.946078431372549,
      "source": "whisper"
    },
    {
      "start": 290.04,
      "end": 291.38,
      "text": "变成语音",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.332143119211947,
      "no_speech_prob": 2.2597562709947283e-11,
      "compression_ratio": 1.946078431372549,
      "source": "whisper"
    },
    {
      "start": 291.38,
      "end": 294.26,
      "text": "对吧",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.332143119211947,
      "no_speech_prob": 2.2597562709947283e-11,
      "compression_ratio": 1.946078431372549,
      "source": "whisper"
    },
    {
      "start": 294.26,
      "end": 295.5,
      "text": "是不是",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.332143119211947,
      "no_speech_prob": 2.2597562709947283e-11,
      "compression_ratio": 1.946078431372549,
      "source": "whisper"
    },
    {
      "start": 295.5,
      "end": 296.42,
      "text": "好",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.332143119211947,
      "no_speech_prob": 2.2597562709947283e-11,
      "compression_ratio": 1.946078431372549,
      "source": "whisper"
    },
    {
      "start": 296.42,
      "end": 297.86,
      "text": "那还有棒连",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.332143119211947,
      "no_speech_prob": 2.2597562709947283e-11,
      "compression_ratio": 1.946078431372549,
      "source": "whisper"
    },
    {
      "start": 297.86,
      "end": 299.18,
      "text": "这个棒连",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.332143119211947,
      "no_speech_prob": 2.2597562709947283e-11,
      "compression_ratio": 1.946078431372549,
      "source": "whisper"
    },
    {
      "start": 299.18,
      "end": 300.58,
      "text": "这个棒连",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.332143119211947,
      "no_speech_prob": 2.2597562709947283e-11,
      "compression_ratio": 1.946078431372549,
      "source": "whisper"
    },
    {
      "start": 300.58,
      "end": 300.86,
      "text": "这个棒连",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.332143119211947,
      "no_speech_prob": 2.2597562709947283e-11,
      "compression_ratio": 1.946078431372549,
      "source": "whisper"
    },
    {
      "start": 300.86,
      "end": 300.9,
      "text": "这个棒连",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.332143119211947,
      "no_speech_prob": 2.2597562709947283e-11,
      "compression_ratio": 1.946078431372549,
      "source": "whisper"
    },
    {
      "start": 306.9,
      "end": 307.62,
      "text": "实时",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.16244075500887203,
      "no_speech_prob": 2.621591659035527e-11,
      "compression_ratio": 1.3585657370517927,
      "source": "whisper"
    },
    {
      "start": 307.62,
      "end": 309.5,
      "text": "人生实时的一个特点",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.16244075500887203,
      "no_speech_prob": 2.621591659035527e-11,
      "compression_ratio": 1.3585657370517927,
      "source": "whisper"
    },
    {
      "start": 309.5,
      "end": 310.44,
      "text": "好",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.16244075500887203,
      "no_speech_prob": 2.621591659035527e-11,
      "compression_ratio": 1.3585657370517927,
      "source": "whisper"
    },
    {
      "start": 310.44,
      "end": 311.6,
      "text": "我讲完这个之后了",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.16244075500887203,
      "no_speech_prob": 2.621591659035527e-11,
      "compression_ratio": 1.3585657370517927,
      "source": "whisper"
    },
    {
      "start": 311.6,
      "end": 313.16,
      "text": "当我这里没讲详细啊",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.16244075500887203,
      "no_speech_prob": 2.621591659035527e-11,
      "compression_ratio": 1.3585657370517927,
      "source": "whisper"
    },
    {
      "start": 313.16,
      "end": 315.44,
      "text": "我重点是想谈两件事情",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.16244075500887203,
      "no_speech_prob": 2.621591659035527e-11,
      "compression_ratio": 1.3585657370517927,
      "source": "whisper"
    },
    {
      "start": 315.44,
      "end": 317.12,
      "text": "第一件事情",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.16244075500887203,
      "no_speech_prob": 2.621591659035527e-11,
      "compression_ratio": 1.3585657370517927,
      "source": "whisper"
    },
    {
      "start": 317.12,
      "end": 319.36,
      "text": "就你既然是一个智能体",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.16244075500887203,
      "no_speech_prob": 2.621591659035527e-11,
      "compression_ratio": 1.3585657370517927,
      "source": "whisper"
    },
    {
      "start": 319.36,
      "end": 322.36,
      "text": "那么智能体得有循环迭代",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.16244075500887203,
      "no_speech_prob": 2.621591659035527e-11,
      "compression_ratio": 1.3585657370517927,
      "source": "whisper"
    },
    {
      "start": 322.36,
      "end": 324.36,
      "text": "得有用户意图识别",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.16244075500887203,
      "no_speech_prob": 2.621591659035527e-11,
      "compression_ratio": 1.3585657370517927,
      "source": "whisper"
    },
    {
      "start": 324.36,
      "end": 326.5,
      "text": "得有工具的调用",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.16244075500887203,
      "no_speech_prob": 2.621591659035527e-11,
      "compression_ratio": 1.3585657370517927,
      "source": "whisper"
    },
    {
      "start": 326.5,
      "end": 327.84,
      "text": "skill的调用",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.16244075500887203,
      "no_speech_prob": 2.621591659035527e-11,
      "compression_ratio": 1.3585657370517927,
      "source": "whisper"
    },
    {
      "start": 327.84,
      "end": 329.44,
      "text": "等等上下文管理",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.16244075500887203,
      "no_speech_prob": 2.621591659035527e-11,
      "compression_ratio": 1.3585657370517927,
      "source": "whisper"
    },
    {
      "start": 329.44,
      "end": 330.9,
      "text": "你既然是个智能体",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.16244075500887203,
      "no_speech_prob": 2.621591659035527e-11,
      "compression_ratio": 1.3585657370517927,
      "source": "whisper"
    },
    {
      "start": 330.9,
      "end": 332.76,
      "text": "你是不是这些都应该有",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.16244075500887203,
      "no_speech_prob": 2.621591659035527e-11,
      "compression_ratio": 1.3585657370517927,
      "source": "whisper"
    },
    {
      "start": 332.76,
      "end": 334.56,
      "text": "没错吧",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.16244075500887203,
      "no_speech_prob": 2.621591659035527e-11,
      "compression_ratio": 1.3585657370517927,
      "source": "whisper"
    },
    {
      "start": 334.56,
      "end": 335.74,
      "text": "好",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.16244075500887203,
      "no_speech_prob": 2.621591659035527e-11,
      "compression_ratio": 1.3585657370517927,
      "source": "whisper"
    },
    {
      "start": 335.74,
      "end": 336.94,
      "text": "但问题在于",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.11469713155774103,
      "no_speech_prob": 1.4856121541284573e-11,
      "compression_ratio": 1.330827067669173,
      "source": "whisper"
    },
    {
      "start": 336.94,
      "end": 340.66,
      "text": "如果你采用这种原生的实时的",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.11469713155774103,
      "no_speech_prob": 1.4856121541284573e-11,
      "compression_ratio": 1.330827067669173,
      "source": "whisper"
    },
    {
      "start": 340.66,
      "end": 341.94,
      "text": "这种价格的方式的时候",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.11469713155774103,
      "no_speech_prob": 1.4856121541284573e-11,
      "compression_ratio": 1.330827067669173,
      "source": "whisper"
    },
    {
      "start": 341.94,
      "end": 344.2,
      "text": "你发现你去找一个这样的模型",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.11469713155774103,
      "no_speech_prob": 1.4856121541284573e-11,
      "compression_ratio": 1.330827067669173,
      "source": "whisper"
    },
    {
      "start": 344.2,
      "end": 345.3,
      "text": "可能能找到",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.11469713155774103,
      "no_speech_prob": 1.4856121541284573e-11,
      "compression_ratio": 1.330827067669173,
      "source": "whisper"
    },
    {
      "start": 345.3,
      "end": 349.34,
      "text": "但是他对智能体的支持不是特别好",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.11469713155774103,
      "no_speech_prob": 1.4856121541284573e-11,
      "compression_ratio": 1.330827067669173,
      "source": "whisper"
    },
    {
      "start": 349.34,
      "end": 351.1,
      "text": "他在智能体",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.11469713155774103,
      "no_speech_prob": 1.4856121541284573e-11,
      "compression_ratio": 1.330827067669173,
      "source": "whisper"
    },
    {
      "start": 351.1,
      "end": 354.7,
      "text": "这个这种数据平台上的测试得分很低",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.11469713155774103,
      "no_speech_prob": 1.4856121541284573e-11,
      "compression_ratio": 1.330827067669173,
      "source": "whisper"
    },
    {
      "start": 354.7,
      "end": 357.58,
      "text": "最近大家听到两个新闻",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.11469713155774103,
      "no_speech_prob": 1.4856121541284573e-11,
      "compression_ratio": 1.330827067669173,
      "source": "whisper"
    },
    {
      "start": 357.58,
      "end": 361.38,
      "text": "第一个新闻是中国上海的一个模型",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.11469713155774103,
      "no_speech_prob": 1.4856121541284573e-11,
      "compression_ratio": 1.330827067669173,
      "source": "whisper"
    },
    {
      "start": 361.38,
      "end": 364.36,
      "text": "叫来自于这个KIMI的K3",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.11469713155774103,
      "no_speech_prob": 1.4856121541284573e-11,
      "compression_ratio": 1.330827067669173,
      "source": "whisper"
    },
    {
      "start": 364.36,
      "end": 364.92,
      "text": "是吧",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.11469713155774103,
      "no_speech_prob": 1.4856121541284573e-11,
      "compression_ratio": 1.330827067669173,
      "source": "whisper"
    },
    {
      "start": 364.92,
      "end": 368.32,
      "text": "还有这个质朴的最新的5.2",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.18389862873515145,
      "no_speech_prob": 1.3242056756679332e-11,
      "compression_ratio": 1.208,
      "source": "whisper"
    },
    {
      "start": 368.32,
      "end": 375.04,
      "text": "在国外所有这些非常知名的测试平台上",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.18389862873515145,
      "no_speech_prob": 1.3242056756679332e-11,
      "compression_ratio": 1.208,
      "source": "whisper"
    },
    {
      "start": 375.04,
      "end": 378.12,
      "text": "得到的分数是非常接近于",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.18389862873515145,
      "no_speech_prob": 1.3242056756679332e-11,
      "compression_ratio": 1.208,
      "source": "whisper"
    },
    {
      "start": 378.12,
      "end": 380.84,
      "text": "OpenAI的5.5对吧",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.18389862873515145,
      "no_speech_prob": 1.3242056756679332e-11,
      "compression_ratio": 1.208,
      "source": "whisper"
    },
    {
      "start": 380.84,
      "end": 384.06,
      "text": "也非常接近于Antherapyc里面的",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.18389862873515145,
      "no_speech_prob": 1.3242056756679332e-11,
      "compression_ratio": 1.208,
      "source": "whisper"
    },
    {
      "start": 384.06,
      "end": 388.46,
      "text": "各种4.6的这个这个Opus的4.6的这样的模型",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.18389862873515145,
      "no_speech_prob": 1.3242056756679332e-11,
      "compression_ratio": 1.208,
      "source": "whisper"
    },
    {
      "start": 388.46,
      "end": 390.4,
      "text": "那大家最重要的你知道吗",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.18389862873515145,
      "no_speech_prob": 1.3242056756679332e-11,
      "compression_ratio": 1.208,
      "source": "whisper"
    },
    {
      "start": 390.4,
      "end": 392.12,
      "text": "是看两个数据得分",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.18389862873515145,
      "no_speech_prob": 1.3242056756679332e-11,
      "compression_ratio": 1.208,
      "source": "whisper"
    },
    {
      "start": 392.12,
      "end": 394.16,
      "text": "一个是什么数据",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.18389862873515145,
      "no_speech_prob": 1.3242056756679332e-11,
      "compression_ratio": 1.208,
      "source": "whisper"
    },
    {
      "start": 394.16,
      "end": 395.94,
      "text": "一个是coding的数据",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.12318358770230921,
      "no_speech_prob": 2.2719857245556696e-11,
      "compression_ratio": 1.5419580419580419,
      "source": "whisper"
    },
    {
      "start": 395.94,
      "end": 397.36,
      "text": "一个是智能体的数据",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.12318358770230921,
      "no_speech_prob": 2.2719857245556696e-11,
      "compression_ratio": 1.5419580419580419,
      "source": "whisper"
    },
    {
      "start": 397.36,
      "end": 400.76,
      "text": "我不知道大家是否能听得懂我讲的话",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.12318358770230921,
      "no_speech_prob": 2.2719857245556696e-11,
      "compression_ratio": 1.5419580419580419,
      "source": "whisper"
    },
    {
      "start": 400.76,
      "end": 404.88,
      "text": "我这里面我今天没有准备好准备好那个图片",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.12318358770230921,
      "no_speech_prob": 2.2719857245556696e-11,
      "compression_ratio": 1.5419580419580419,
      "source": "whisper"
    },
    {
      "start": 404.88,
      "end": 409.32,
      "text": "我这个本来按道理说给大家要看那个图片的描述了",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.12318358770230921,
      "no_speech_prob": 2.2719857245556696e-11,
      "compression_ratio": 1.5419580419580419,
      "source": "whisper"
    },
    {
      "start": 409.32,
      "end": 411.0,
      "text": "我可能可能这个质谱里面",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.12318358770230921,
      "no_speech_prob": 2.2719857245556696e-11,
      "compression_ratio": 1.5419580419580419,
      "source": "whisper"
    },
    {
      "start": 411.0,
      "end": 412.18,
      "text": "这个摩塔社区里面有",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.12318358770230921,
      "no_speech_prob": 2.2719857245556696e-11,
      "compression_ratio": 1.5419580419580419,
      "source": "whisper"
    },
    {
      "start": 412.18,
      "end": 413.84,
      "text": "对摩塔社区里面有",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.12318358770230921,
      "no_speech_prob": 2.2719857245556696e-11,
      "compression_ratio": 1.5419580419580419,
      "source": "whisper"
    },
    {
      "start": 413.84,
      "end": 416.44,
      "text": "一般来说一个模型的新模型发布",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.12318358770230921,
      "no_speech_prob": 2.2719857245556696e-11,
      "compression_ratio": 1.5419580419580419,
      "source": "whisper"
    },
    {
      "start": 416.44,
      "end": 420.8,
      "text": "它都会有这样的一个各个数据平台的一个得分情况",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.12318358770230921,
      "no_speech_prob": 2.2719857245556696e-11,
      "compression_ratio": 1.5419580419580419,
      "source": "whisper"
    },
    {
      "start": 420.8,
      "end": 421.4,
      "text": "好吧",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.12318358770230921,
      "no_speech_prob": 2.2719857245556696e-11,
      "compression_ratio": 1.5419580419580419,
      "source": "whisper"
    },
    {
      "start": 421.4,
      "end": 423.38,
      "text": "我看一下有没有",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.12318358770230921,
      "no_speech_prob": 2.2719857245556696e-11,
      "compression_ratio": 1.5419580419580419,
      "source": "whisper"
    },
    {
      "start": 423.38,
      "end": 425.44,
      "text": "这个大家听懂的",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.17023125672951722,
      "no_speech_prob": 1.6819838924431174e-11,
      "compression_ratio": 1.5256916996047432,
      "source": "whisper"
    },
    {
      "start": 425.44,
      "end": 426.12,
      "text": "Q1啊",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.17023125672951722,
      "no_speech_prob": 1.6819838924431174e-11,
      "compression_ratio": 1.5256916996047432,
      "source": "whisper"
    },
    {
      "start": 426.12,
      "end": 429.06,
      "text": "就是任何一个牛逼的模型",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.17023125672951722,
      "no_speech_prob": 1.6819838924431174e-11,
      "compression_ratio": 1.5256916996047432,
      "source": "whisper"
    },
    {
      "start": 429.06,
      "end": 431.88,
      "text": "之所以在全国来说",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.17023125672951722,
      "no_speech_prob": 1.6819838924431174e-11,
      "compression_ratio": 1.5256916996047432,
      "source": "whisper"
    },
    {
      "start": 431.88,
      "end": 433.66,
      "text": "是比较高的一个",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.17023125672951722,
      "no_speech_prob": 1.6819838924431174e-11,
      "compression_ratio": 1.5256916996047432,
      "source": "whisper"
    },
    {
      "start": 433.66,
      "end": 435.74,
      "text": "比较比较高知名度的",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.17023125672951722,
      "no_speech_prob": 1.6819838924431174e-11,
      "compression_ratio": 1.5256916996047432,
      "source": "whisper"
    },
    {
      "start": 435.74,
      "end": 437.26,
      "text": "他有两个分数很重要",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.17023125672951722,
      "no_speech_prob": 1.6819838924431174e-11,
      "compression_ratio": 1.5256916996047432,
      "source": "whisper"
    },
    {
      "start": 437.26,
      "end": 438.36,
      "text": "一个是这个分数",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.17023125672951722,
      "no_speech_prob": 1.6819838924431174e-11,
      "compression_ratio": 1.5256916996047432,
      "source": "whisper"
    },
    {
      "start": 438.36,
      "end": 438.82,
      "text": "大家看",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.17023125672951722,
      "no_speech_prob": 1.6819838924431174e-11,
      "compression_ratio": 1.5256916996047432,
      "source": "whisper"
    },
    {
      "start": 438.82,
      "end": 440.08,
      "text": "这个分数",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.17023125672951722,
      "no_speech_prob": 1.6819838924431174e-11,
      "compression_ratio": 1.5256916996047432,
      "source": "whisper"
    },
    {
      "start": 440.08,
      "end": 441.5,
      "text": "好吧",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.17023125672951722,
      "no_speech_prob": 1.6819838924431174e-11,
      "compression_ratio": 1.5256916996047432,
      "source": "whisper"
    },
    {
      "start": 441.5,
      "end": 444.7,
      "text": "还有一个是关于coding的那个分数",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.17023125672951722,
      "no_speech_prob": 1.6819838924431174e-11,
      "compression_ratio": 1.5256916996047432,
      "source": "whisper"
    },
    {
      "start": 444.7,
      "end": 445.86,
      "text": "但最正的是这个",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.17023125672951722,
      "no_speech_prob": 1.6819838924431174e-11,
      "compression_ratio": 1.5256916996047432,
      "source": "whisper"
    },
    {
      "start": 445.86,
      "end": 448.58,
      "text": "大家为什么要重点看这个指标吗",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.17023125672951722,
      "no_speech_prob": 1.6819838924431174e-11,
      "compression_ratio": 1.5256916996047432,
      "source": "whisper"
    },
    {
      "start": 448.58,
      "end": 451.04,
      "text": "看这个基准测试的数据",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.17023125672951722,
      "no_speech_prob": 1.6819838924431174e-11,
      "compression_ratio": 1.5256916996047432,
      "source": "whisper"
    },
    {
      "start": 451.04,
      "end": 453.28,
      "text": "这个基准测试的数据集的测试结果",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.17023125672951722,
      "no_speech_prob": 1.6819838924431174e-11,
      "compression_ratio": 1.5256916996047432,
      "source": "whisper"
    },
    {
      "start": 453.28,
      "end": 454.72,
      "text": "测试结果的分数吗",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1536178069860757,
      "no_speech_prob": 1.8665610723722637e-11,
      "compression_ratio": 1.541501976284585,
      "source": "whisper"
    },
    {
      "start": 454.72,
      "end": 455.9,
      "text": "为什么大家看这个",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1536178069860757,
      "no_speech_prob": 1.8665610723722637e-11,
      "compression_ratio": 1.541501976284585,
      "source": "whisper"
    },
    {
      "start": 455.9,
      "end": 456.48,
      "text": "大家知道不",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1536178069860757,
      "no_speech_prob": 1.8665610723722637e-11,
      "compression_ratio": 1.541501976284585,
      "source": "whisper"
    },
    {
      "start": 456.48,
      "end": 458.14,
      "text": "就不用我再说了吧",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1536178069860757,
      "no_speech_prob": 1.8665610723722637e-11,
      "compression_ratio": 1.541501976284585,
      "source": "whisper"
    },
    {
      "start": 458.14,
      "end": 460.16,
      "text": "好吧",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1536178069860757,
      "no_speech_prob": 1.8665610723722637e-11,
      "compression_ratio": 1.541501976284585,
      "source": "whisper"
    },
    {
      "start": 460.16,
      "end": 462.04,
      "text": "大家为什么要看这个指标",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1536178069860757,
      "no_speech_prob": 1.8665610723722637e-11,
      "compression_ratio": 1.541501976284585,
      "source": "whisper"
    },
    {
      "start": 462.04,
      "end": 463.68,
      "text": "其他的不看",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1536178069860757,
      "no_speech_prob": 1.8665610723722637e-11,
      "compression_ratio": 1.541501976284585,
      "source": "whisper"
    },
    {
      "start": 463.68,
      "end": 467.06,
      "text": "或者其他的很少看",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1536178069860757,
      "no_speech_prob": 1.8665610723722637e-11,
      "compression_ratio": 1.541501976284585,
      "source": "whisper"
    },
    {
      "start": 467.06,
      "end": 468.52,
      "text": "不是完全不看啊",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1536178069860757,
      "no_speech_prob": 1.8665610723722637e-11,
      "compression_ratio": 1.541501976284585,
      "source": "whisper"
    },
    {
      "start": 468.52,
      "end": 471.06,
      "text": "其实我们跟国外的模型比较",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1536178069860757,
      "no_speech_prob": 1.8665610723722637e-11,
      "compression_ratio": 1.541501976284585,
      "source": "whisper"
    },
    {
      "start": 471.06,
      "end": 473.32,
      "text": "重要的就比较两个最重要的指标",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1536178069860757,
      "no_speech_prob": 1.8665610723722637e-11,
      "compression_ratio": 1.541501976284585,
      "source": "whisper"
    },
    {
      "start": 473.32,
      "end": 475.42,
      "text": "一个是这个智能体的分数",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1536178069860757,
      "no_speech_prob": 1.8665610723722637e-11,
      "compression_ratio": 1.541501976284585,
      "source": "whisper"
    },
    {
      "start": 475.42,
      "end": 476.38,
      "text": "智能体的能力",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1536178069860757,
      "no_speech_prob": 1.8665610723722637e-11,
      "compression_ratio": 1.541501976284585,
      "source": "whisper"
    },
    {
      "start": 476.38,
      "end": 478.2,
      "text": "智能体的准确性的能力",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1536178069860757,
      "no_speech_prob": 1.8665610723722637e-11,
      "compression_ratio": 1.541501976284585,
      "source": "whisper"
    },
    {
      "start": 478.2,
      "end": 479.86,
      "text": "还有一个是一个coding能力",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1536178069860757,
      "no_speech_prob": 1.8665610723722637e-11,
      "compression_ratio": 1.541501976284585,
      "source": "whisper"
    },
    {
      "start": 479.86,
      "end": 480.44,
      "text": "就这两个",
      "chunk": 0,
      "language": "zh",
      "avg_logprob": -0.1536178069860757,
      "no_speech_prob": 1.8665610723722637e-11,
      "compression_ratio": 1.541501976284585,
      "source": "whisper"
    },
    {
      "start": 481.244,
      "end": 492.564,
      "text": "你发现,你像DeepSync V4,Flash 0731,这个正式版,你发现他是不是非常接近于Opus Anthropic的Opus 4.8,对吧?",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.23638980946642288,
      "no_speech_prob": 9.38413236006852e-12,
      "compression_ratio": 1.2818791946308725,
      "source": "whisper"
    },
    {
      "start": 492.564,
      "end": 494.324,
      "text": "没错吧",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.22200365887572435,
      "no_speech_prob": 1.8696310125076998e-11,
      "compression_ratio": 1.23,
      "source": "whisper"
    },
    {
      "start": 494.324,
      "end": 497.444,
      "text": "所以我们国外也好",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.22200365887572435,
      "no_speech_prob": 1.8696310125076998e-11,
      "compression_ratio": 1.23,
      "source": "whisper"
    },
    {
      "start": 497.444,
      "end": 497.944,
      "text": "国内也好",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.22200365887572435,
      "no_speech_prob": 1.8696310125076998e-11,
      "compression_ratio": 1.23,
      "source": "whisper"
    },
    {
      "start": 497.944,
      "end": 498.644,
      "text": "大家都知道",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.22200365887572435,
      "no_speech_prob": 1.8696310125076998e-11,
      "compression_ratio": 1.23,
      "source": "whisper"
    },
    {
      "start": 498.644,
      "end": 502.504,
      "text": "这个DeepSync V4的正式版本",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.22200365887572435,
      "no_speech_prob": 1.8696310125076998e-11,
      "compression_ratio": 1.23,
      "source": "whisper"
    },
    {
      "start": 502.504,
      "end": 505.144,
      "text": "好像跟这个Opus 4.8",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.22200365887572435,
      "no_speech_prob": 1.8696310125076998e-11,
      "compression_ratio": 1.23,
      "source": "whisper"
    },
    {
      "start": 505.144,
      "end": 505.684,
      "text": "是相接近",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.22200365887572435,
      "no_speech_prob": 1.8696310125076998e-11,
      "compression_ratio": 1.23,
      "source": "whisper"
    },
    {
      "start": 505.684,
      "end": 506.764,
      "text": "其实不单是DeepSync",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.22200365887572435,
      "no_speech_prob": 1.8696310125076998e-11,
      "compression_ratio": 1.23,
      "source": "whisper"
    },
    {
      "start": 506.764,
      "end": 509.064,
      "text": "还有那个刚刚我说的Kimin的Key3",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.22200365887572435,
      "no_speech_prob": 1.8696310125076998e-11,
      "compression_ratio": 1.23,
      "source": "whisper"
    },
    {
      "start": 509.064,
      "end": 511.184,
      "text": "以及智朴的最新发布的那个模型",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.22200365887572435,
      "no_speech_prob": 1.8696310125076998e-11,
      "compression_ratio": 1.23,
      "source": "whisper"
    },
    {
      "start": 511.184,
      "end": 511.724,
      "text": "是吧",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.22200365887572435,
      "no_speech_prob": 1.8696310125076998e-11,
      "compression_ratio": 1.23,
      "source": "whisper"
    },
    {
      "start": 511.724,
      "end": 512.644,
      "text": "对吧",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.22200365887572435,
      "no_speech_prob": 1.8696310125076998e-11,
      "compression_ratio": 1.23,
      "source": "whisper"
    },
    {
      "start": 512.644,
      "end": 515.144,
      "text": "那我那老师你讲这话什么意思呢",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.22200365887572435,
      "no_speech_prob": 1.8696310125076998e-11,
      "compression_ratio": 1.23,
      "source": "whisper"
    },
    {
      "start": 515.144,
      "end": 516.504,
      "text": "是这样子的",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.22200365887572435,
      "no_speech_prob": 1.8696310125076998e-11,
      "compression_ratio": 1.23,
      "source": "whisper"
    },
    {
      "start": 516.504,
      "end": 517.744,
      "text": "就是你的智能",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.22200365887572435,
      "no_speech_prob": 1.8696310125076998e-11,
      "compression_ratio": 1.23,
      "source": "whisper"
    },
    {
      "start": 517.744,
      "end": 521.424,
      "text": "你要你要想作为智能体的核心推敌模型",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.22200365887572435,
      "no_speech_prob": 1.8696310125076998e-11,
      "compression_ratio": 1.23,
      "source": "whisper"
    },
    {
      "start": 521.424,
      "end": 523.384,
      "text": "那么你一定要找",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.2746694260749264,
      "no_speech_prob": 1.6821122619803397e-11,
      "compression_ratio": 1.47136563876652,
      "source": "whisper"
    },
    {
      "start": 523.384,
      "end": 525.144,
      "text": "你一定要去找",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.2746694260749264,
      "no_speech_prob": 1.6821122619803397e-11,
      "compression_ratio": 1.47136563876652,
      "source": "whisper"
    },
    {
      "start": 525.144,
      "end": 528.664,
      "text": "这个分数比较高的那些模型",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.2746694260749264,
      "no_speech_prob": 1.6821122619803397e-11,
      "compression_ratio": 1.47136563876652,
      "source": "whisper"
    },
    {
      "start": 528.664,
      "end": 531.364,
      "text": "懂意思吧",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.2746694260749264,
      "no_speech_prob": 1.6821122619803397e-11,
      "compression_ratio": 1.47136563876652,
      "source": "whisper"
    },
    {
      "start": 531.364,
      "end": 532.284,
      "text": "现在我明白了吗",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.2746694260749264,
      "no_speech_prob": 1.6821122619803397e-11,
      "compression_ratio": 1.47136563876652,
      "source": "whisper"
    },
    {
      "start": 532.284,
      "end": 534.224,
      "text": "那么这个分数比较高的那些模型",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.2746694260749264,
      "no_speech_prob": 1.6821122619803397e-11,
      "compression_ratio": 1.47136563876652,
      "source": "whisper"
    },
    {
      "start": 534.224,
      "end": 534.864,
      "text": "都是什么",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.2746694260749264,
      "no_speech_prob": 1.6821122619803397e-11,
      "compression_ratio": 1.47136563876652,
      "source": "whisper"
    },
    {
      "start": 534.864,
      "end": 535.464,
      "text": "这个",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.2746694260749264,
      "no_speech_prob": 1.6821122619803397e-11,
      "compression_ratio": 1.47136563876652,
      "source": "whisper"
    },
    {
      "start": 535.464,
      "end": 536.444,
      "text": "这个",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.2746694260749264,
      "no_speech_prob": 1.6821122619803397e-11,
      "compression_ratio": 1.47136563876652,
      "source": "whisper"
    },
    {
      "start": 536.444,
      "end": 537.384,
      "text": "这个",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.2746694260749264,
      "no_speech_prob": 1.6821122619803397e-11,
      "compression_ratio": 1.47136563876652,
      "source": "whisper"
    },
    {
      "start": 537.384,
      "end": 538.064,
      "text": "这个",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.2746694260749264,
      "no_speech_prob": 1.6821122619803397e-11,
      "compression_ratio": 1.47136563876652,
      "source": "whisper"
    },
    {
      "start": 538.064,
      "end": 538.564,
      "text": "是吧",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.2746694260749264,
      "no_speech_prob": 1.6821122619803397e-11,
      "compression_ratio": 1.47136563876652,
      "source": "whisper"
    },
    {
      "start": 538.564,
      "end": 540.284,
      "text": "当然还有这个OpenAI的",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.2746694260749264,
      "no_speech_prob": 1.6821122619803397e-11,
      "compression_ratio": 1.47136563876652,
      "source": "whisper"
    },
    {
      "start": 540.284,
      "end": 542.184,
      "text": "这个GPT5.5 5.6",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.2746694260749264,
      "no_speech_prob": 1.6821122619803397e-11,
      "compression_ratio": 1.47136563876652,
      "source": "whisper"
    },
    {
      "start": 542.184,
      "end": 542.844,
      "text": "那个",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.2746694260749264,
      "no_speech_prob": 1.6821122619803397e-11,
      "compression_ratio": 1.47136563876652,
      "source": "whisper"
    },
    {
      "start": 542.844,
      "end": 543.384,
      "text": "那个",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.2746694260749264,
      "no_speech_prob": 1.6821122619803397e-11,
      "compression_ratio": 1.47136563876652,
      "source": "whisper"
    },
    {
      "start": 543.384,
      "end": 543.704,
      "text": "那个",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.2746694260749264,
      "no_speech_prob": 1.6821122619803397e-11,
      "compression_ratio": 1.47136563876652,
      "source": "whisper"
    },
    {
      "start": 543.704,
      "end": 545.264,
      "text": "他没列在这里头",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.2746694260749264,
      "no_speech_prob": 1.6821122619803397e-11,
      "compression_ratio": 1.47136563876652,
      "source": "whisper"
    },
    {
      "start": 545.264,
      "end": 546.544,
      "text": "对吧",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.2746694260749264,
      "no_speech_prob": 1.6821122619803397e-11,
      "compression_ratio": 1.47136563876652,
      "source": "whisper"
    },
    {
      "start": 546.544,
      "end": 547.484,
      "text": "哦",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.2746694260749264,
      "no_speech_prob": 1.6821122619803397e-11,
      "compression_ratio": 1.47136563876652,
      "source": "whisper"
    },
    {
      "start": 547.484,
      "end": 548.104,
      "text": "明白了",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.2746694260749264,
      "no_speech_prob": 1.6821122619803397e-11,
      "compression_ratio": 1.47136563876652,
      "source": "whisper"
    },
    {
      "start": 548.104,
      "end": 549.504,
      "text": "老师少说大概你懂意思了",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.2746694260749264,
      "no_speech_prob": 1.6821122619803397e-11,
      "compression_ratio": 1.47136563876652,
      "source": "whisper"
    },
    {
      "start": 549.504,
      "end": 549.964,
      "text": "就是说",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.2746694260749264,
      "no_speech_prob": 1.6821122619803397e-11,
      "compression_ratio": 1.47136563876652,
      "source": "whisper"
    },
    {
      "start": 549.964,
      "end": 552.904,
      "text": "我如果要开发一个流逼的智能体",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13500730614913137,
      "no_speech_prob": 1.489813654387273e-11,
      "compression_ratio": 1.3745019920318724,
      "source": "whisper"
    },
    {
      "start": 552.904,
      "end": 555.604,
      "text": "那我智能体里面的核心推理模型",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13500730614913137,
      "no_speech_prob": 1.489813654387273e-11,
      "compression_ratio": 1.3745019920318724,
      "source": "whisper"
    },
    {
      "start": 555.604,
      "end": 557.444,
      "text": "和核心的意图识别模型",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13500730614913137,
      "no_speech_prob": 1.489813654387273e-11,
      "compression_ratio": 1.3745019920318724,
      "source": "whisper"
    },
    {
      "start": 557.444,
      "end": 558.944,
      "text": "得是很强的",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13500730614913137,
      "no_speech_prob": 1.489813654387273e-11,
      "compression_ratio": 1.3745019920318724,
      "source": "whisper"
    },
    {
      "start": 558.944,
      "end": 560.904,
      "text": "而且怎么证明它强",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13500730614913137,
      "no_speech_prob": 1.489813654387273e-11,
      "compression_ratio": 1.3745019920318724,
      "source": "whisper"
    },
    {
      "start": 560.904,
      "end": 562.204,
      "text": "就是看这个模型",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13500730614913137,
      "no_speech_prob": 1.489813654387273e-11,
      "compression_ratio": 1.3745019920318724,
      "source": "whisper"
    },
    {
      "start": 562.204,
      "end": 566.644,
      "text": "在这个数据级上的基准测试的得分来判断的",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13500730614913137,
      "no_speech_prob": 1.489813654387273e-11,
      "compression_ratio": 1.3745019920318724,
      "source": "whisper"
    },
    {
      "start": 566.644,
      "end": 567.344,
      "text": "是吧",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13500730614913137,
      "no_speech_prob": 1.489813654387273e-11,
      "compression_ratio": 1.3745019920318724,
      "source": "whisper"
    },
    {
      "start": 567.344,
      "end": 567.864,
      "text": "是的",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13500730614913137,
      "no_speech_prob": 1.489813654387273e-11,
      "compression_ratio": 1.3745019920318724,
      "source": "whisper"
    },
    {
      "start": 567.864,
      "end": 568.904,
      "text": "没错是这个意思",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13500730614913137,
      "no_speech_prob": 1.489813654387273e-11,
      "compression_ratio": 1.3745019920318724,
      "source": "whisper"
    },
    {
      "start": 568.904,
      "end": 569.384,
      "text": "好",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13500730614913137,
      "no_speech_prob": 1.489813654387273e-11,
      "compression_ratio": 1.3745019920318724,
      "source": "whisper"
    },
    {
      "start": 569.384,
      "end": 571.004,
      "text": "接下来我再讲另外一句话",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13500730614913137,
      "no_speech_prob": 1.489813654387273e-11,
      "compression_ratio": 1.3745019920318724,
      "source": "whisper"
    },
    {
      "start": 571.004,
      "end": 571.564,
      "text": "好",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13500730614913137,
      "no_speech_prob": 1.489813654387273e-11,
      "compression_ratio": 1.3745019920318724,
      "source": "whisper"
    },
    {
      "start": 571.564,
      "end": 573.684,
      "text": "那有一些",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13500730614913137,
      "no_speech_prob": 1.489813654387273e-11,
      "compression_ratio": 1.3745019920318724,
      "source": "whisper"
    },
    {
      "start": 573.684,
      "end": 577.024,
      "text": "有一些全模态的大模型",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13500730614913137,
      "no_speech_prob": 1.489813654387273e-11,
      "compression_ratio": 1.3745019920318724,
      "source": "whisper"
    },
    {
      "start": 577.024,
      "end": 581.764,
      "text": "他虽然也能够支持语音的输入",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.221840763092041,
      "no_speech_prob": 1.4164798674276557e-11,
      "compression_ratio": 1.2735849056603774,
      "source": "whisper"
    },
    {
      "start": 581.764,
      "end": 583.904,
      "text": "同时也支持语音的输出",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.221840763092041,
      "no_speech_prob": 1.4164798674276557e-11,
      "compression_ratio": 1.2735849056603774,
      "source": "whisper"
    },
    {
      "start": 583.904,
      "end": 586.084,
      "text": "但问题在于",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.221840763092041,
      "no_speech_prob": 1.4164798674276557e-11,
      "compression_ratio": 1.2735849056603774,
      "source": "whisper"
    },
    {
      "start": 586.084,
      "end": 590.144,
      "text": "他在这个数据级的基准测试的得分",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.221840763092041,
      "no_speech_prob": 1.4164798674276557e-11,
      "compression_ratio": 1.2735849056603774,
      "source": "whisper"
    },
    {
      "start": 590.144,
      "end": 592.864,
      "text": "是非常非常的低的",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.221840763092041,
      "no_speech_prob": 1.4164798674276557e-11,
      "compression_ratio": 1.2735849056603774,
      "source": "whisper"
    },
    {
      "start": 592.864,
      "end": 594.124,
      "text": "举个例子",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.221840763092041,
      "no_speech_prob": 1.4164798674276557e-11,
      "compression_ratio": 1.2735849056603774,
      "source": "whisper"
    },
    {
      "start": 594.124,
      "end": 596.944,
      "text": "比如说像全模态大模型",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.221840763092041,
      "no_speech_prob": 1.4164798674276557e-11,
      "compression_ratio": 1.2735849056603774,
      "source": "whisper"
    },
    {
      "start": 596.944,
      "end": 597.964,
      "text": "有一个",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.221840763092041,
      "no_speech_prob": 1.4164798674276557e-11,
      "compression_ratio": 1.2735849056603774,
      "source": "whisper"
    },
    {
      "start": 597.964,
      "end": 600.704,
      "text": "就是我们经常讲的",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.221840763092041,
      "no_speech_prob": 1.4164798674276557e-11,
      "compression_ratio": 1.2735849056603774,
      "source": "whisper"
    },
    {
      "start": 600.704,
      "end": 602.364,
      "text": "签文里面有一个",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.221840763092041,
      "no_speech_prob": 1.4164798674276557e-11,
      "compression_ratio": 1.2735849056603774,
      "source": "whisper"
    },
    {
      "start": 602.364,
      "end": 604.384,
      "text": "稍等啊",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.221840763092041,
      "no_speech_prob": 1.4164798674276557e-11,
      "compression_ratio": 1.2735849056603774,
      "source": "whisper"
    },
    {
      "start": 604.384,
      "end": 605.004,
      "text": "搜索一下",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.221840763092041,
      "no_speech_prob": 1.4164798674276557e-11,
      "compression_ratio": 1.2735849056603774,
      "source": "whisper"
    },
    {
      "start": 605.004,
      "end": 607.064,
      "text": "看是能搜到啊",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.2427072727933843,
      "no_speech_prob": 1.4794216199320864e-11,
      "compression_ratio": 1.3392857142857142,
      "source": "whisper"
    },
    {
      "start": 607.064,
      "end": 608.244,
      "text": "这个搜索框去了",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.2427072727933843,
      "no_speech_prob": 1.4794216199320864e-11,
      "compression_ratio": 1.3392857142857142,
      "source": "whisper"
    },
    {
      "start": 608.244,
      "end": 617.604,
      "text": "这个",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.2427072727933843,
      "no_speech_prob": 1.4794216199320864e-11,
      "compression_ratio": 1.3392857142857142,
      "source": "whisper"
    },
    {
      "start": 617.604,
      "end": 619.384,
      "text": "好吧",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.2427072727933843,
      "no_speech_prob": 1.4794216199320864e-11,
      "compression_ratio": 1.3392857142857142,
      "source": "whisper"
    },
    {
      "start": 619.384,
      "end": 623.604,
      "text": "这就是千万里面一个非常典型的全模态大模型",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.2427072727933843,
      "no_speech_prob": 1.4794216199320864e-11,
      "compression_ratio": 1.3392857142857142,
      "source": "whisper"
    },
    {
      "start": 623.604,
      "end": 625.184,
      "text": "什么叫全模态大模型呢",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.2427072727933843,
      "no_speech_prob": 1.4794216199320864e-11,
      "compression_ratio": 1.3392857142857142,
      "source": "whisper"
    },
    {
      "start": 625.184,
      "end": 626.064,
      "text": "就是",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.2427072727933843,
      "no_speech_prob": 1.4794216199320864e-11,
      "compression_ratio": 1.3392857142857142,
      "source": "whisper"
    },
    {
      "start": 626.064,
      "end": 628.824,
      "text": "这个网页打不开啊",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.2427072727933843,
      "no_speech_prob": 1.4794216199320864e-11,
      "compression_ratio": 1.3392857142857142,
      "source": "whisper"
    },
    {
      "start": 628.824,
      "end": 631.124,
      "text": "就是他输入的可以是语音",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.2427072727933843,
      "no_speech_prob": 1.4794216199320864e-11,
      "compression_ratio": 1.3392857142857142,
      "source": "whisper"
    },
    {
      "start": 631.124,
      "end": 632.784,
      "text": "然后他可以思考",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.2427072727933843,
      "no_speech_prob": 1.4794216199320864e-11,
      "compression_ratio": 1.3392857142857142,
      "source": "whisper"
    },
    {
      "start": 632.784,
      "end": 633.504,
      "text": "思考完之后",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13948758864244878,
      "no_speech_prob": 1.9833709327676807e-11,
      "compression_ratio": 1.403225806451613,
      "source": "whisper"
    },
    {
      "start": 633.504,
      "end": 634.624,
      "text": "它输出的结果",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13948758864244878,
      "no_speech_prob": 1.9833709327676807e-11,
      "compression_ratio": 1.403225806451613,
      "source": "whisper"
    },
    {
      "start": 634.624,
      "end": 635.424,
      "text": "也可以是语音",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13948758864244878,
      "no_speech_prob": 1.9833709327676807e-11,
      "compression_ratio": 1.403225806451613,
      "source": "whisper"
    },
    {
      "start": 635.424,
      "end": 636.624,
      "text": "也可以是文字",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13948758864244878,
      "no_speech_prob": 1.9833709327676807e-11,
      "compression_ratio": 1.403225806451613,
      "source": "whisper"
    },
    {
      "start": 636.624,
      "end": 637.784,
      "text": "语音文字都支持",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13948758864244878,
      "no_speech_prob": 1.9833709327676807e-11,
      "compression_ratio": 1.403225806451613,
      "source": "whisper"
    },
    {
      "start": 637.784,
      "end": 638.744,
      "text": "那这种模型",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13948758864244878,
      "no_speech_prob": 1.9833709327676807e-11,
      "compression_ratio": 1.403225806451613,
      "source": "whisper"
    },
    {
      "start": 638.744,
      "end": 639.944,
      "text": "是不是非常适合",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13948758864244878,
      "no_speech_prob": 1.9833709327676807e-11,
      "compression_ratio": 1.403225806451613,
      "source": "whisper"
    },
    {
      "start": 639.944,
      "end": 640.704,
      "text": "我们做这种",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13948758864244878,
      "no_speech_prob": 1.9833709327676807e-11,
      "compression_ratio": 1.403225806451613,
      "source": "whisper"
    },
    {
      "start": 640.704,
      "end": 642.544,
      "text": "这个语音的智能体",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13948758864244878,
      "no_speech_prob": 1.9833709327676807e-11,
      "compression_ratio": 1.403225806451613,
      "source": "whisper"
    },
    {
      "start": 642.544,
      "end": 642.844,
      "text": "是的",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13948758864244878,
      "no_speech_prob": 1.9833709327676807e-11,
      "compression_ratio": 1.403225806451613,
      "source": "whisper"
    },
    {
      "start": 642.844,
      "end": 644.144,
      "text": "非常适合的",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13948758864244878,
      "no_speech_prob": 1.9833709327676807e-11,
      "compression_ratio": 1.403225806451613,
      "source": "whisper"
    },
    {
      "start": 644.144,
      "end": 645.224,
      "text": "但问题在于",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13948758864244878,
      "no_speech_prob": 1.9833709327676807e-11,
      "compression_ratio": 1.403225806451613,
      "source": "whisper"
    },
    {
      "start": 645.224,
      "end": 646.184,
      "text": "这样的模型",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13948758864244878,
      "no_speech_prob": 1.9833709327676807e-11,
      "compression_ratio": 1.403225806451613,
      "source": "whisper"
    },
    {
      "start": 646.184,
      "end": 647.544,
      "text": "它在智能体",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13948758864244878,
      "no_speech_prob": 1.9833709327676807e-11,
      "compression_ratio": 1.403225806451613,
      "source": "whisper"
    },
    {
      "start": 647.544,
      "end": 650.264,
      "text": "这个数据级上的评测得分",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13948758864244878,
      "no_speech_prob": 1.9833709327676807e-11,
      "compression_ratio": 1.403225806451613,
      "source": "whisper"
    },
    {
      "start": 650.264,
      "end": 652.084,
      "text": "是远远低于",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13948758864244878,
      "no_speech_prob": 1.9833709327676807e-11,
      "compression_ratio": 1.403225806451613,
      "source": "whisper"
    },
    {
      "start": 652.084,
      "end": 653.104,
      "text": "大家看到的",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13948758864244878,
      "no_speech_prob": 1.9833709327676807e-11,
      "compression_ratio": 1.403225806451613,
      "source": "whisper"
    },
    {
      "start": 653.104,
      "end": 656.044,
      "text": "这些主流的推定模型",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13948758864244878,
      "no_speech_prob": 1.9833709327676807e-11,
      "compression_ratio": 1.403225806451613,
      "source": "whisper"
    },
    {
      "start": 656.044,
      "end": 659.084,
      "text": "现在明白意思吧",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13948758864244878,
      "no_speech_prob": 1.9833709327676807e-11,
      "compression_ratio": 1.403225806451613,
      "source": "whisper"
    },
    {
      "start": 659.084,
      "end": 660.324,
      "text": "所以",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13948758864244878,
      "no_speech_prob": 1.9833709327676807e-11,
      "compression_ratio": 1.403225806451613,
      "source": "whisper"
    },
    {
      "start": 660.324,
      "end": 664.364,
      "text": "在实际的企业开发过程中",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.1341407692950705,
      "no_speech_prob": 1.2462792083056673e-11,
      "compression_ratio": 1.2941176470588236,
      "source": "whisper"
    },
    {
      "start": 664.364,
      "end": 667.024,
      "text": "如果要做一个语音模型",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.1341407692950705,
      "no_speech_prob": 1.2462792083056673e-11,
      "compression_ratio": 1.2941176470588236,
      "source": "whisper"
    },
    {
      "start": 667.024,
      "end": 669.944,
      "text": "你又想要保证实时性",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.1341407692950705,
      "no_speech_prob": 1.2462792083056673e-11,
      "compression_ratio": 1.2941176470588236,
      "source": "whisper"
    },
    {
      "start": 669.944,
      "end": 675.784,
      "text": "你又想保证我这个智能体核心的推理模型和意图识别模型",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.1341407692950705,
      "no_speech_prob": 1.2462792083056673e-11,
      "compression_ratio": 1.2941176470588236,
      "source": "whisper"
    },
    {
      "start": 675.784,
      "end": 677.304,
      "text": "是很好的模型",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.1341407692950705,
      "no_speech_prob": 1.2462792083056673e-11,
      "compression_ratio": 1.2941176470588236,
      "source": "whisper"
    },
    {
      "start": 677.304,
      "end": 678.324,
      "text": "是很牛的模型",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.1341407692950705,
      "no_speech_prob": 1.2462792083056673e-11,
      "compression_ratio": 1.2941176470588236,
      "source": "whisper"
    },
    {
      "start": 678.324,
      "end": 682.264,
      "text": "那这两个要求都要满足",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.1341407692950705,
      "no_speech_prob": 1.2462792083056673e-11,
      "compression_ratio": 1.2941176470588236,
      "source": "whisper"
    },
    {
      "start": 682.264,
      "end": 683.504,
      "text": "说实话",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.1341407692950705,
      "no_speech_prob": 1.2462792083056673e-11,
      "compression_ratio": 1.2941176470588236,
      "source": "whisper"
    },
    {
      "start": 683.504,
      "end": 685.644,
      "text": "很难有完美的方案",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.1341407692950705,
      "no_speech_prob": 1.2462792083056673e-11,
      "compression_ratio": 1.2941176470588236,
      "source": "whisper"
    },
    {
      "start": 685.644,
      "end": 691.044,
      "text": "所以为了同时满足这两个要求",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13760869126570852,
      "no_speech_prob": 1.714505447336645e-11,
      "compression_ratio": 1.2614840989399294,
      "source": "whisper"
    },
    {
      "start": 691.044,
      "end": 694.844,
      "text": "我们今天重新",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13760869126570852,
      "no_speech_prob": 1.714505447336645e-11,
      "compression_ratio": 1.2614840989399294,
      "source": "whisper"
    },
    {
      "start": 694.844,
      "end": 697.184,
      "text": "根据当了我前两天也是看到网上论文",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13760869126570852,
      "no_speech_prob": 1.714505447336645e-11,
      "compression_ratio": 1.2614840989399294,
      "source": "whisper"
    },
    {
      "start": 697.184,
      "end": 698.364,
      "text": "根据网上论文",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13760869126570852,
      "no_speech_prob": 1.714505447336645e-11,
      "compression_ratio": 1.2614840989399294,
      "source": "whisper"
    },
    {
      "start": 698.364,
      "end": 700.024,
      "text": "我构建了一个新的方案",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13760869126570852,
      "no_speech_prob": 1.714505447336645e-11,
      "compression_ratio": 1.2614840989399294,
      "source": "whisper"
    },
    {
      "start": 700.024,
      "end": 701.904,
      "text": "叫三明治方案",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13760869126570852,
      "no_speech_prob": 1.714505447336645e-11,
      "compression_ratio": 1.2614840989399294,
      "source": "whisper"
    },
    {
      "start": 701.904,
      "end": 703.504,
      "text": "那三明治方案它的区别是",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13760869126570852,
      "no_speech_prob": 1.714505447336645e-11,
      "compression_ratio": 1.2614840989399294,
      "source": "whisper"
    },
    {
      "start": 703.504,
      "end": 704.824,
      "text": "它的特点是什么呢",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13760869126570852,
      "no_speech_prob": 1.714505447336645e-11,
      "compression_ratio": 1.2614840989399294,
      "source": "whisper"
    },
    {
      "start": 704.824,
      "end": 707.604,
      "text": "就是重点放在智能体这一块",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13760869126570852,
      "no_speech_prob": 1.714505447336645e-11,
      "compression_ratio": 1.2614840989399294,
      "source": "whisper"
    },
    {
      "start": 707.604,
      "end": 709.484,
      "text": "智能体的核心推理模型",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13760869126570852,
      "no_speech_prob": 1.714505447336645e-11,
      "compression_ratio": 1.2614840989399294,
      "source": "whisper"
    },
    {
      "start": 709.484,
      "end": 711.044,
      "text": "我还是使用这些像",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13760869126570852,
      "no_speech_prob": 1.714505447336645e-11,
      "compression_ratio": 1.2614840989399294,
      "source": "whisper"
    },
    {
      "start": 711.044,
      "end": 712.804,
      "text": "DeepSync V4的正式版",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13760869126570852,
      "no_speech_prob": 1.714505447336645e-11,
      "compression_ratio": 1.2614840989399294,
      "source": "whisper"
    },
    {
      "start": 712.804,
      "end": 715.184,
      "text": "或者Opus的4.8",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.13760869126570852,
      "no_speech_prob": 1.714505447336645e-11,
      "compression_ratio": 1.2614840989399294,
      "source": "whisper"
    },
    {
      "start": 715.184,
      "end": 717.904,
      "text": "或者是OpenAI的5.6",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.18739146656460232,
      "no_speech_prob": 1.6280518599920413e-11,
      "compression_ratio": 1.2840909090909092,
      "source": "whisper"
    },
    {
      "start": 717.904,
      "end": 719.024,
      "text": "5.5这样的模型",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.18739146656460232,
      "no_speech_prob": 1.6280518599920413e-11,
      "compression_ratio": 1.2840909090909092,
      "source": "whisper"
    },
    {
      "start": 719.024,
      "end": 721.204,
      "text": "或者是质朴的",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.18739146656460232,
      "no_speech_prob": 1.6280518599920413e-11,
      "compression_ratio": 1.2840909090909092,
      "source": "whisper"
    },
    {
      "start": 721.204,
      "end": 722.544,
      "text": "对吧最新的模型",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.18739146656460232,
      "no_speech_prob": 1.6280518599920413e-11,
      "compression_ratio": 1.2840909090909092,
      "source": "whisper"
    },
    {
      "start": 722.544,
      "end": 724.824,
      "text": "好然后呢",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.18739146656460232,
      "no_speech_prob": 1.6280518599920413e-11,
      "compression_ratio": 1.2840909090909092,
      "source": "whisper"
    },
    {
      "start": 724.824,
      "end": 725.484,
      "text": "但这两个",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.18739146656460232,
      "no_speech_prob": 1.6280518599920413e-11,
      "compression_ratio": 1.2840909090909092,
      "source": "whisper"
    },
    {
      "start": 725.484,
      "end": 728.524,
      "text": "他在智能体这方面",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.18739146656460232,
      "no_speech_prob": 1.6280518599920413e-11,
      "compression_ratio": 1.2840909090909092,
      "source": "whisper"
    },
    {
      "start": 728.524,
      "end": 729.584,
      "text": "在智能体这个",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.18739146656460232,
      "no_speech_prob": 1.6280518599920413e-11,
      "compression_ratio": 1.2840909090909092,
      "source": "whisper"
    },
    {
      "start": 729.584,
      "end": 733.244,
      "text": "这个基准测试数据上表现分数非常高",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.18739146656460232,
      "no_speech_prob": 1.6280518599920413e-11,
      "compression_ratio": 1.2840909090909092,
      "source": "whisper"
    },
    {
      "start": 733.244,
      "end": 734.844,
      "text": "对吧还不错非常好",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.18739146656460232,
      "no_speech_prob": 1.6280518599920413e-11,
      "compression_ratio": 1.2840909090909092,
      "source": "whisper"
    },
    {
      "start": 734.844,
      "end": 738.684,
      "text": "可是问题是他对语音的输入和输出",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.18739146656460232,
      "no_speech_prob": 1.6280518599920413e-11,
      "compression_ratio": 1.2840909090909092,
      "source": "whisper"
    },
    {
      "start": 738.684,
      "end": 739.484,
      "text": "不是特别支持",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.18739146656460232,
      "no_speech_prob": 1.6280518599920413e-11,
      "compression_ratio": 1.2840909090909092,
      "source": "whisper"
    },
    {
      "start": 739.484,
      "end": 740.684,
      "text": "对吧",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.18739146656460232,
      "no_speech_prob": 1.6280518599920413e-11,
      "compression_ratio": 1.2840909090909092,
      "source": "whisper"
    },
    {
      "start": 740.684,
      "end": 743.164,
      "text": "所以就有一个这么一个矛盾点",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.18739146656460232,
      "no_speech_prob": 1.6280518599920413e-11,
      "compression_ratio": 1.2840909090909092,
      "source": "whisper"
    },
    {
      "start": 743.164,
      "end": 743.844,
      "text": "那怎么办呢",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.18739146656460232,
      "no_speech_prob": 1.6280518599920413e-11,
      "compression_ratio": 1.2840909090909092,
      "source": "whisper"
    },
    {
      "start": 743.844,
      "end": 745.104,
      "text": "我在它的前面",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.21087125631479117,
      "no_speech_prob": 1.4486458040086081e-11,
      "compression_ratio": 1.1555555555555554,
      "source": "whisper"
    },
    {
      "start": 745.104,
      "end": 746.484,
      "text": "加上一个STD模型",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.21087125631479117,
      "no_speech_prob": 1.4486458040086081e-11,
      "compression_ratio": 1.1555555555555554,
      "source": "whisper"
    },
    {
      "start": 746.484,
      "end": 747.204,
      "text": "我在它的后面",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.21087125631479117,
      "no_speech_prob": 1.4486458040086081e-11,
      "compression_ratio": 1.1555555555555554,
      "source": "whisper"
    },
    {
      "start": 747.204,
      "end": 748.364,
      "text": "加上一个TTS模型",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.21087125631479117,
      "no_speech_prob": 1.4486458040086081e-11,
      "compression_ratio": 1.1555555555555554,
      "source": "whisper"
    },
    {
      "start": 748.364,
      "end": 749.504,
      "text": "把它夹到中间",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.21087125631479117,
      "no_speech_prob": 1.4486458040086081e-11,
      "compression_ratio": 1.1555555555555554,
      "source": "whisper"
    },
    {
      "start": 749.504,
      "end": 750.544,
      "text": "就称之为这样",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.21087125631479117,
      "no_speech_prob": 1.4486458040086081e-11,
      "compression_ratio": 1.1555555555555554,
      "source": "whisper"
    },
    {
      "start": 750.544,
      "end": 751.484,
      "text": "三米之架构",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.21087125631479117,
      "no_speech_prob": 1.4486458040086081e-11,
      "compression_ratio": 1.1555555555555554,
      "source": "whisper"
    },
    {
      "start": 751.484,
      "end": 752.864,
      "text": "现在我应该讲清楚了",
      "chunk": 1,
      "language": "zh",
      "avg_logprob": -0.21087125631479117,
      "no_speech_prob": 1.4486458040086081e-11,
      "compression_ratio": 1.1555555555555554,
      "source": "whisper"
    }
  ],
  "transcript_selection": {
    "schema": "videonote.transcript_selection.v1",
    "status": "SELECTED",
    "requested_language": "zh",
    "duration": 756.41325,
    "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": 366,
      "multilingual": false,
      "first_start": 0.0,
      "final_end": 752.864,
      "duration_coverage": 0.995308,
      "maximum_gap_seconds": 6.0,
      "inverted_count": 0,
      "language": {
        "requested": "zh",
        "matched": true,
        "latin": 255,
        "han": 2901,
        "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": 756.413,
      "rejected_segment_count": 0,
      "rejected_segments": []
    }
  }
}