A robot can chat through voice.(version 1.0) made by apis.
It simplely make a sentence-to-sentence conversion so far.
It will updated in real-time speech-robot.
语音聊天机器人，通过调用api组合而成。
目前只能实现通过单句问答。
后续将会改成实时对话。
Python 3.0+, Pygame, Pyaudio, wave
apply for api_key,secret_key,app_id in baidu voice.
apply for api_key in turing robot.
Write these into config.py, and run main.py.
申请百度语音的api_key,secrek_key,app_id
申请图灵机器人的api_key
将其填写在config文件当中，运行main.py即可。
The chat robot is designed by five components.
这个聊天机器人通过五个组件构成。
This file record the human voice and transfer it into ask.wav to recognize.
记录人声，并且将其保存为ask.wav以便后续的识别。
This file used baidu speech recognizer to recognize voice.
调用百度语音识别，来识别人声内容。
This file used turing chat-robot to get answer text.
通过图灵机器人获得回答内容文本。
This file make the text into speech and save as ans.mp3 in local.
通过百度语音合成将文本内容合成语音并保存在 ans.mp3.
This filere play the ans.mp3, but it has some bugs.
Python has few modules to play .mp3 on Mac. So I use pygame/mixer.
If you use windows, you can edit this file to get a fluent voice.
播放ans.mp3文件，但是这个有一点问题，python在mac上能够播放MP3的库很少。
这里使用了Pygame的组件来播放，稍微有点失真。
如果你使用windows系统，可以更改这个文件以便获得更流利的发音。