关于语音识别功能的使用

[ShowUNow]

前置确认

1. 网络能够访问openai接口 #351
2. python 已安装：版本在 3.7 ~ 3.10 之间，依赖已安装
3. 在已有 issue 中未搜索到类似问题
4. FAQS 中无类似问题

问题描述

"voice_reply_voice": true这个语句没在config-template.json里找到，不过应该是添加到chatgpt-on-wechat/config.json里面吧？...

对应语音合成平台的key是阿里云的

终端日志 (如有报错)

环境

• 操作系统类型 (Mac/Windows/Linux)：Windows 10
• Python版本 ( 执行 python3 -V )：python 3.9
• pip版本 ( 依赖问题此项必填，执行 pip3 -V)：pip 23.0.1

[dhlsam]

同问，如何添加语音合成平台的api key

[zhayujie]

一、接收语音，回复文字

config.json 中设置 "speech_recognition": true 即可， 这个配置在 template中有。 启动程序后 私聊 就会支持语音识别，会对所有收到的语音消息转成文字，并回复文字。

二、接收语音，回复语音

有多种可选平台，以下提供两种方案，默认是OpenAI识别，Google语音合成：

1.百度

(1) 安装依赖： pip3 install baidu-aip chardet

(2) 在 config.json 中添加配置：

{
    "speech_recognition": true,
    "voice_reply_voice": true,
    "baidu_app_id": "YOUR BAIDU APP ID",
    "baidu_api_key": "YOUR BAIDU API KEY",
    "baidu_secret_key": "YOUR BAIDU SERVICE KEY"
}

2.Google

(1) 安装 SpeechRecognition： pip3 install SpeechRecognition

(2) 安装 ffmpeg 和 espeak

• MacOS: brew install ffmpeg espeak
• Windows: 下载ffmpeg.exe
• Linux: apt-get install ffmpeg espeak

(3) 在 config.json 中添加配置：

{
    "speech_recognition": true,
    "voice_reply_voice": true
}

注：在itchat中只能回复mp3文件；在wechaty中可以回复语音消息，wechaty中使用需要安装额外依赖：

pip3 install pysilk-mod
pip3 install pydub

三、群组语音

"group_speech_recognition": true加入配置项

PR：
#385 by @wanggang1987
#476 by @Chiaki-Chan
#623 by @Chiaki-Chan

[4331599]

你好。为什么我填好了之后还是提示这个错误呢？
[ERROR][2023-03-10 09:04:23][wechat_channel.py:151] - 'session_id' Traceback (most recent call last): File "E:\Enviroment\PythonWorks\src\chatgpt-on-wechat-20230310\chatgpt-on-wechat-1.0.5\channel\wechat\wechat_channel.py", line 145, in _do_send_voice reply_text = super().build_reply_content(query, context) File "E:\Enviroment\PythonWorks\src\chatgpt-on-wechat-20230310\chatgpt-on-wechat-1.0.5\channel\channel.py", line 31, in build_reply_content return Bridge().fetch_reply_content(query, context) File "E:\Enviroment\PythonWorks\src\chatgpt-on-wechat-20230310\chatgpt-on-wechat-1.0.5\bridge\bridge.py", line 10, in fetch_reply_content return bot_factory.create_bot("chatGPT").reply(query, context) File "E:\Enviroment\PythonWorks\src\chatgpt-on-wechat-20230310\chatgpt-on-wechat-1.0.5\bot\chatgpt\chat_gpt_bot.py", line 27, in reply session_id = context['session_id'] KeyError: 'session_id'

[zhayujie]

你好。为什么我填好了之后还是提示这个错误呢？ [ERROR][2023-03-10 09:04:23][wechat_channel.py:151] - 'session_id' Traceback (most recent call last): File "E:\Enviroment\PythonWorks\src\chatgpt-on-wechat-20230310\chatgpt-on-wechat-1.0.5\channel\wechat\wechat_channel.py", line 145, in _do_send_voice reply_text = super().build_reply_content(query, context) File "E:\Enviroment\PythonWorks\src\chatgpt-on-wechat-20230310\chatgpt-on-wechat-1.0.5\channel\channel.py", line 31, in build_reply_content return Bridge().fetch_reply_content(query, context) File "E:\Enviroment\PythonWorks\src\chatgpt-on-wechat-20230310\chatgpt-on-wechat-1.0.5\bridge\bridge.py", line 10, in fetch_reply_content return bot_factory.create_bot("chatGPT").reply(query, context) File "E:\Enviroment\PythonWorks\src\chatgpt-on-wechat-20230310\chatgpt-on-wechat-1.0.5\bot\chatgpt\chat_gpt_bot.py", line 27, in reply session_id = context['session_id'] KeyError: 'session_id'

修改了下，可以获取最新代码再试试

[yuyang0702]

我同样的问题，获取最新的代码后在执行，提示:
[ERROR][2023-03-10 10:00:50][wechat_channel.py:151] - No module named 'aip'
Traceback (most recent call last):
File "/root/chatgpt-on-wechat/channel/wechat/wechat_channel.py", line 147, in _do_send_voice
replyFile = super().build_text_to_voice(reply_text)
File "/root/chatgpt-on-wechat/channel/channel.py", line 37, in build_text_to_voice
return Bridge().fetch_text_to_voice(text)
File "/root/chatgpt-on-wechat/bridge/bridge.py", line 16, in fetch_text_to_voice
return voice_factory.create_voice("baidu").textToVoice(text)
File "/root/chatgpt-on-wechat/voice/voice_factory.py", line 12, in create_voice
from voice.baidu.baidu_voice import BaiduVoice
File "/root/chatgpt-on-wechat/voice/baidu/baidu_voice.py", line 6, in
from aip import AipSpeech
ModuleNotFoundError: No module named 'aip'

[Ayuyyae]

我同样的问题，获取最新的代码后在执行，提示: [ERROR][2023-03-10 10:00:50][wechat_channel.py:151] - No module named 'aip' Traceback (most recent call last): File "/root/chatgpt-on-wechat/channel/wechat/wechat_channel.py", line 147, in _do_send_voice replyFile = super().build_text_to_voice(reply_text) File "/root/chatgpt-on-wechat/channel/channel.py", line 37, in build_text_to_voice return Bridge().fetch_text_to_voice(text) File "/root/chatgpt-on-wechat/bridge/bridge.py", line 16, in fetch_text_to_voice return voice_factory.create_voice("baidu").textToVoice(text) File "/root/chatgpt-on-wechat/voice/voice_factory.py", line 12, in create_voice from voice.baidu.baidu_voice import BaiduVoice File "/root/chatgpt-on-wechat/voice/baidu/baidu_voice.py", line 6, in from aip import AipSpeech ModuleNotFoundError: No module named 'aip'

你需要安装：pip install baidu-aip

[wanggang1987]

安装下pip3 install baidu-aip chardet

[huanyan]

因为调用了baidu-aip库，终端输入pip3 install baidu-aip chardet 。

[yuyang0702]

谢谢，已经解决了，程序默认的语音类型的百度，我没安装百度的这个库，刚刚改成谷歌了。谢谢~~

[wangjiangmin1115]

[INFO][2023-03-10 16:30:44][openai_voice.py:22] - [Openai] voiceToText text=回复一个语言 voice file name=tmp/230310-163043.mp3
[INFO][2023-03-10 16:30:44][chat_gpt_bot.py:26] - [OPEN_AI] query=回复一个语言
[INFO][2023-03-10 16:30:46][google_voice.py:49] - [Google] textToVoice text=您好！请问题用哪种语言提问，我将用相同的语言进行回答。 voice file name=tmp/语音回复_1678437046.mp3
[INFO][2023-03-10 16:30:46][wechat_channel.py:149] - [WX] sendFile=tmp/语音回复_1678437046.mp3, receiver=@68bd8f5f7464a0ffee8f9ef10bf401e8b83abc33ad375cc6c2c6167d6842e19b

使用的是谷歌的语音类型，但是微信没有回复任何消息

[wangjiangmin1115]

在itchat中只能回复mp3文件，在wechaty中可以回复语音消息
wechaty 所需要的版本是多少，需要指定某个版本吗？

itchat和wechaty，是在哪个配置项，还是哪个文件上指定了

[lingwecho]

wechaty比itchat更容易被封吗？