Zero-shot voice cloning docs - where?
Is there a guide on how to do zero-shot voice cloning with Orpheus? I'm not finding anything proper.
Is there a guide on how to do zero-shot voice cloning with Orpheus? I'm not finding anything proper.
exactly, cannot find code, and no doc. How should we even use that functionality?
I found something in the Issues section on GitHub, but I also found people saying it was broken / inconsistent - and replies saying they're working on it. I've set an alert if that gets updated. Until then, this is off my radar.
I found something in the Issues section on GitHub, but I also found people saying it was broken / inconsistent - and replies saying they're working on it. I've set an alert if that gets updated. Until then, this is off my radar.
What is other best voice clone model?
There was something but I can't remember. Squirrel?? Damn, something like that. Check the open Issues on GitHub, was someone talking about what else they've tried. Right now I'm using ElevenLabs - their zero-shot Instance Voice Cloning is actually pretty good, but their TTS rates are expensive. Keeping a look out for alternatives. I don't need yet - still at poc stage, but at scale it makes sense to find something else. Until then a good abstraction layer is the key.
OpenVoice does voice cloning
@MercyfulKing - thanks! :)