讯飞云语音的Python实现



只是简单的用Python调用dll，自己用着觉得还不错，分享给大家。



双击test.bat就可以实现对wav_test.wav的识别，这个wav是sdk demmo中的123456789的那个。

其它语音文件的话，可以仿照test文件自己写一个。



双击test2.bat可以识别麦克风的输入。这个需要pyaudio的支持。



我是在win7的CMD下运行正常，但是xp的CMD可能会乱码~~~因为日志输出的是unicode编码的文字。



使用方法可以参看src里面的源码，需要注意的是：请把代码里appid=zzzzzzzz替换为自己的AppID



By wangqy(wangqyfm@foxmail.com)

2013-1-26-Python implementation of information Fei voice simply call in Python dll, they used feel pretty good, for everyone to share. Double test.bat wav_test.wav can be achieved on the recognition that this is sdk demmo wav that the 123456789. Other audio files, then you can write a test file modeled. Double test2.bat microphone input can be identified. This requires pyaudio support. I was in the CMD win7 run properly, but the xp CMD may be garbled ~ ~ ~ Because log output is unicode encoded text. Using methods can be found in src inside source, the need to note that: Please code appid = zzzzzzzz replace with your own AppID By wangqy (wangqyfm@foxmail.com) 2013-1-26