- added more config options to CoquiEngine constructor - support for metal shaders gpu acceleration (you may need to set the environment variable `PYTORCH_ENABLE_MPS_FALLBACK=1` to use the CPU as a fallback for ops not implemented in mps yet) - default voice added in case no wav or speaker latents json file present
0.2.1
General stability improvements around Coqui XTTS 2.0 synthesis
0.2.0
Coqui XTTS 2.0 support
0.1.9
- Upgrade to PyTorch 2.1.0 - Support of new OpenAI API - Support of Coqui XTTS 1.1