We need to get this somewhere between XTTS and Kokoro
I love XTTS for voice cloning, emotion, and "newness" within it's voice creations, but it just does not have cleanliness and crispness in it, that's required for TTS.
Kokoro however, while it doesn't have nearly as much expression, it's much more actually listenable.
I wonder if continued work could combine the positives of them both. Looking forward to seeing how it plays out, and I'm contributing the community in my own ways (thought they're not TTS, it's related) or I'd contribute myself.
Working on that
Working on that
Working on that permissively, or with a CC-BY-NC license? 😉
Working on that
Would love to help with this and stay updated. Any ways I can contribute?
Hey, sure! You can DM me on Discord, “yukiarimo#4577,” and I can add you to our dev group. Here’s the WIP repo, btw: https://github.com/yukiarimo/yunatts
Just sent you a Discord request!
Working on that
Very excited! I'd love to join your discord, do you have a group as well?
Sure, DM me and I’ll add you!