This adds the ability to use the Coqui XTTS model which is based on Tortoise but highly optimized. It allows you to supply a short sample voice for cloning, and then does an amazing job of copying that voice. GPU absolutely required, and even then it's pretty slow (but worth it!)
This adds the ability to use the Coqui XTTS model which is based on Tortoise but highly optimized. It allows you to supply a short sample voice for cloning, and then does an amazing job of copying that voice. GPU absolutely required, and even then it's pretty slow (but worth it!)