🐸💬 - a deep learning toolkit for Text-to-Speech, battle-tested in research and production
Full Changelog: https://github.com/coqui-ai/TTS/compare/v0.20.0...v0.20.1
make style
& re-enable it in CI by @akx in https://github.com/coqui-ai/TTS/pull/3127
Full Changelog: https://github.com/coqui-ai/TTS/compare/v0.19.1...v0.20.0
Full Changelog: https://github.com/coqui-ai/TTS/compare/v0.19.0...v0.19.1
Full Changelog: https://github.com/coqui-ai/TTS/compare/v0.18.2...v0.19.0
Full Changelog: https://github.com/coqui-ai/TTS/compare/v0.18.1...v0.18.2
Full Changelog: https://github.com/coqui-ai/TTS/compare/v0.18.0...v0.18.1
Full Changelog: https://github.com/coqui-ai/TTS/compare/v0.17.10...v0.18.0
This model is trained on top of XTTS v1, using output masking. We mask the part of the output that is used as the audio prompt while training and don't compute loss for that segment. This helps us to resolve the hallucination issue that V1 experienced.
ne_hifigan
that was trained without denoising that brought some EQ and compression profile that might be unwanted for some use-casesFull Changelog: https://github.com/coqui-ai/TTS/compare/v0.17.9...v0.17.10
Full Changelog: https://github.com/coqui-ai/TTS/compare/v0.17.8...v0.17.9
Full Changelog: https://github.com/coqui-ai/TTS/compare/v0.17.7...v0.17.8