Closed k-aito closed 6 months ago
Yes, cloning with XTTS is supported.
https://github.com/aedocw/epub2tts?tab=readme-ov-file#xtts-using-your-own-voice-clone
You can run epub2tts my-book.txt --xtts voice-1.wav,voice-2.wav,voice-3.wav
and if your three files (voice-1.wav, voice-2.wav and voice-3.wav) are wave files with the voice you want to clone, XTTS will try to clone that speaker.
If you were asking a different question though, please let me know. I'm not sure what you mean by tts_to_file
.
Hello @aedocw
Thanks, shame on me, I didn't notice it on the main page.
In the coqui-tts documentation they use tts_to_file
with a speaker_wav, but it's not needed anymore because everything is already documented. I just needed to read the README better ^^
Thanks again for this software that works really good in my case.
No problem, there are so many different options with epub2tts, it's not easy to recognize them all at first glance. Glad you are enjoying it, and feel free to open an issue if you run into problems!
Hello
Thanks for your software, it's really working well. I was curious if it would be possible to add cloning with xtts. It works normally speaker_wav with tts, but you use another API if I understood the code well.
I will try to make a PR to use
tts_to_file
but I think that the audio enhancement at the end would maybe be problematic, so if you have a better idea, I take it ^^Cheers