neosapience/editts

How to use voice files instead pure TTS?

Vadim2S opened this issue · 4 comments

Vadim2S commented

In papers you say about LJ speech dataset test (4.3 Content replacement). Can you provide code for loading voice files instead pure sample generation in tts.py?

jaketae commented

Hello @Vadim2S, thanks for opening this issue, and apologies for the belated reply. The modeling code is a direct adaptation of Grad-TTS, so you can refer to the upstream repository for detailed instructions on how to load data. Hope this helps!

mvoodarla commented

I have this question as well. Looking at the inference code, it is unclear how I could drop-in replace running Grad-TTS with my own source WAV file. Any tips would be appreciated :) Or if you have pointers to any other methods of doing something similar.

jaketae commented

Hey @mvoodarla, thanks for opening this issue.

how I could drop-in replace running Grad-TTS with my own source WAV file

Could you explain in more detail what you mean by this? I assume this is in the context of content replacement.

mvoodarla commented

Yes, I would like to have the ability to submit a source wav file with or without a text transcription of it, and be able to replace a word or a set of words that was said in that source wav file. Does that make sense?