search

yl4579 / AuxiliaryASR

Joint CTC-S2S Phoneme-level ASR for Voice Conversion and TTS (Text-Mel Alignment)
MIT License
111 stars 30 forks source link

Is there anyone who has used the phonemizer? Any advice, please, on how to change the code correctly #8

Closed ahmeftah closed 1 year ago

yl4579 commented 1 year ago

You may refer to https://github.com/yl4579/StyleTTS/blob/main/meldataset.py as it uses phonemizer.

ziyaad30 commented 1 year ago

You may refer to https://github.com/yl4579/StyleTTS/blob/main/meldataset.py as it uses phonemizer.

That is all you refer someone after you changed the code from StyleTTS and projects and said nothing nor update either way. This is lame

yl4579 commented 1 year ago

@ziyaad30 I'm sorry but this is exactly what you need to change for phonezmiers. All you need to do is change the number of tokens to 178 and use the text cleaner in that meldataset.py file. Your text also needs to be already preprocessed by phonemizer so they are in the IPA form. I will try to change this repo for phonemizer later when I get more time because I'm very busy with another project.