Diamondfan / CTC_pytorch

CTC end -to-end ASR for timit and 863 corpus.
218 stars 48 forks source link

Decode wav file into phonemes? #2

Closed Terrance82 closed 4 years ago

Terrance82 commented 5 years ago

Hi, I'm interested in your project however I'm new in Speech recognition etc. I want to know if your program able to convert from wav audio files and detect its phoneme?

If so, how do i proceed on doing so? Thank you

Diamondfan commented 5 years ago

@Terrance82 The project can only convert wav files to phoneme with an offline mode. It means you have to extract features in advance and get some ark files. Then run test.py to get phoneme sequence after you train a model with enough data.