知识卡片

普通读书笔记卡 · 6794 · 生成模型:GPT-6

语音前后端会引入额外误差

理解等级:中级

语音识别先把声音转文字,核心对话系统处理,再由语音合成输出声音。识别错字会传到意图和槽位,合成语调又影响理解;听起来像人不证明任务能力提高,需分别测延迟、识别和输出可懂度。

这张卡片看不懂?

已有反馈 · 0 条

    还没有人提交反馈。

    在整书笔记中的位置

    原书出处 · 1 处

    • · 查看原书依据

      Giving your chatbot a voice Though we haven’t talked much about voice processing in this book, you may wonder if the NLP tools you’ve learned can help you build a voice assistant similar to Siri or Mycroft. To build a voice assistant, you can preprocess the inputs of your chatbot with existing voice recognition or speech-to-text (STT) software. You can then use speech generation, or

    参考来源

    - 位置:《Natural Language Processing in Action, Second Edition》第12章《对话引擎与会话设计》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Natural Language Processing in Action, Second Edition/_epub-src/OEBPS/Text/chapter-12.html) - 结论依据:摘录中关于语音前后端会引入额外误差的定义、比较或条件支撑卡片;原书表述不严谨的范围已在覆盖记录逐项限定,不把案例结果外推为普遍结论。 - 原始内容:Giving your chatbot a voice Though we haven’t talked much about voice processing in this book, you may wonder if the NLP tools you’ve learned can help you build a voice assistant similar to Siri or Mycroft. To build a voice assistant, you can preprocess the inputs of your chatbot with existing voice recognition or speech-to-text (STT) software. You can then use speech generation, or