Fork of:

https://github.com/Mozer/talk-llama-fast

Refactored to compile with Linux.

The XTTS server is already downloaded here, so when the readme says to download that, no need.

But you will need to download the models. Unfortunately I haven't been able to figure out what the correct names of the models should be, so I haven't been able to run it myself.

If anyone is able to figure out, starting from the state of this repo, what needs to be done to get the models loaded, please let me know. I'm pretty new to this stuff but wanted to get this repo running on Linux, so hopefully this helps someone.

If you want, open an issue and let me know how to run it, and I'll update this readme so myself and others can use it.