I've been using elevenlabs and burning lots of money now regenerating because for some reason my voice is speaking in multiple accents now. Basically with my cloned voice I am looking for something that can be consistent, not conversational like. I have a lot of reference audio. Is it possible to get something identical to what elevenlabs can do? I've tried VOXCPM before and it was decent, I'm thinking of giving it another shot. But I've also heard of Vibevoice. What would you recommend these days when focused on quality to get it almost the same as the reference audio?
3080 12GB VRAM
32 gb of RAM
Any help would be appreciated.
[link] [comments]