Expand description
Coqui XTTS v2 backend.
XTTS v2 is a Coqui-released zero-shot voice-cloning TTS model
supporting 17 languages and ~6-second reference clips. Like
MOSS-TTS this crate hosts the model out-of-process via Python
and talks to it over HTTP. The trait surface + CAPS ship
today; the HTTP client is gated behind xtts-http.