search

litagin02 / Style-Bert-VITS2

Style-Bert-VITS2: Bert-VITS2 with more controllable voice styles.
GNU Affero General Public License v3.0
717 stars 86 forks source link

I am looking for adding "tone" and "emotion/style" in a technology similar to "rvc-tts-edge" (your other repo) #29

Closed APISeeker closed 8 months ago

APISeeker commented 8 months ago

Hello, Is this the right place to continue using rvc-tts-edge but with tone modificaiton and emotions modificaiton? I would like to be able to use these 2 options in the other repositoruy of rvc tts, don't know if it's possible ?

litagin02 commented 8 months ago

rvc-tts-webui uses edge-tts as a source audio of TTS, and we cannot control emotions and tones of ouputs of edge-tts, so I think it's difficult.

APISeeker commented 8 months ago

So we should look for a way to "take edgetts audio -> modify its tone/emotion(style) -> insert it to rvc-tts-edge"?

Wonder if it the step 2 is doable or not. (Modifying tone and style of an existing already audio)

litagin02 commented 8 months ago

I'm not a specialist of such kind of things, so I don't know it's doable or not, and is not related for this repo.

APISeeker commented 8 months ago

OK @litagin02, I have a request, I was not able to geenrate CHINESE audio by inserting chinese text to rvc tts (https://github.com/litagin02/rvc-tts-webui) Could you re-open the repository please? I would like to explore the issues and see if someone had the same issue as me: 

.....\communicate.py", line 458, in save
    with metadata, open(audio_fname, "wb") as audio:
OSError: [Errno 22] Invalid argument:

Thanks in advance. ^^