Fixed a couple of bugs and added TIMIT dataset declaration

jzlianglu / pykaldi2

Yet another speech toolkit based on Kaldi and PyTorch

MIT License

173 stars 33 forks source link

Closed Serhiy-Shekhovtsov closed 4 years ago

Serhiy-Shekhovtsov commented 4 years ago

Couple of fixes I had to make to successfully run the train_ce.py script. Also added the TIMIT dataset class.