search

zzw922cn / Automatic_Speech_Recognition

End-to-end Automatic Speech Recognition for Madarian and English in Tensorflow
MIT License
2.84k stars 539 forks source link

timit_preprocess.py #2

Open duxiansheng opened 7 years ago

duxiansheng commented 7 years ago

I can't read timit, the problem is ValueError: File format 'NIST'... not understood. can you help me?

zzw922cn commented 7 years ago

You should convert the format of your audio first~

brianlan commented 7 years ago

@duxiansheng I use sndfile converter to convert NIST to WAV. Please refer to this repo and their home page. The TIMIT files you downloaded maybe ended with .wav, but they actually are NIST files. So you need to rename them as .nist before using sndfile to do converting.

zzw922cn commented 7 years ago

@duxiansheng @brianlan nst2wav script has been updated just now, it's in the feature folder.

duxiansheng commented 7 years ago

I can covert it with Auducity,but there are a lot of work to do. And is there a way to covert all of .wav by code?

zzw922cn commented 7 years ago

@duxiansheng yes, you can see my new code in timit or libri or wsj, it can automatically process both wav and nist files.