What's Changed

• Drop diffusion from XTTS by @erogol in https://github.com/coqui-ai/TTS/pull/3150
• Bug fixes and add support for multiples speaker references on XTTS inference by @Edresson in https://github.com/coqui-ai/TTS/pull/3149
• Fix XTTS v2.0 training recipe by @Edresson in https://github.com/coqui-ai/TTS/pull/3154

Full Changelog: https://github.com/coqui-ai/TTS/compare/v0.20.0...v0.20.1

What's Changed

• Run make style & re-enable it in CI by @akx in https://github.com/coqui-ai/TTS/pull/3127
• XTTS v2.0 by @Edresson in https://github.com/coqui-ai/TTS/pull/3137

Full Changelog: https://github.com/coqui-ai/TTS/compare/v0.19.1...v0.20.0

What's Changed

• Second round of issue fixing for XTTS v1.1 by @WeberJulian in https://github.com/coqui-ai/TTS/pull/3103
• fix for issue 3067 by @Aya-AlJafari in https://github.com/coqui-ai/TTS/pull/3109
• Bug: self.model_name needed to be initialized. by @vltmedia in https://github.com/coqui-ai/TTS/pull/2983

New Contributors

• @vltmedia made their first contribution in https://github.com/coqui-ai/TTS/pull/2983

Full Changelog: https://github.com/coqui-ai/TTS/compare/v0.19.0...v0.19.1

What's Changed

• XTTS v1.1 GPT Trainer by @Edresson in https://github.com/coqui-ai/TTS/pull/3086

Full Changelog: https://github.com/coqui-ai/TTS/compare/v0.18.2...v0.19.0

What's Changed

• Fix xtts v1.1 by @WeberJulian in https://github.com/coqui-ai/TTS/pull/3096

Full Changelog: https://github.com/coqui-ai/TTS/compare/v0.18.1...v0.18.2

What's Changed

• Bug fix on XTTS v1.1 inference by @Edresson and @WeberJulian in https://github.com/coqui-ai/TTS/pull/3093

Full Changelog: https://github.com/coqui-ai/TTS/compare/v0.18.0...v0.18.1

What's Changed

• XTTS v1.1 by @WeberJulian in https://github.com/coqui-ai/TTS/pull/3089

Full Changelog: https://github.com/coqui-ai/TTS/compare/v0.17.10...v0.18.0

XTTS v1.1

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.

Changes

• Add Japanese
• Resolve the hallucination issue (repeating the audio prompt)
• Increased expressivity
• Hash check to control model version
• Added ne_hifigan that was trained without denoising that brought some EQ and compression profile that might be unwanted for some use-cases

Full Changelog: https://github.com/coqui-ai/TTS/compare/v0.17.9...v0.17.10

What's Changed

• fixed bugs in fastpitch tts synthesis by @Aya-AlJafari in https://github.com/coqui-ai/TTS/pull/3058
• Update AnalyzeDataset.ipynb by @meryemsakin in https://github.com/coqui-ai/TTS/pull/2783
• Synthesizer skips over embeddings file if model only has one speaker by @wonkothesanest in https://github.com/coqui-ai/TTS/pull/2587
• fixed typo of docs\source\implementing_a_new_model.md by @Subash-Lamichhane in https://github.com/coqui-ai/TTS/pull/3066
• fixed typo of /docs by @Subash-Lamichhane in https://github.com/coqui-ai/TTS/pull/3065
• Add play and speed to cli options by @David-bfg in https://github.com/coqui-ai/TTS/pull/3027
• Fix doc dataset by @WeberJulian in https://github.com/coqui-ai/TTS/pull/3070
• fix readme by @WeberJulian in https://github.com/coqui-ai/TTS/pull/3071

New Contributors

• @meryemsakin made their first contribution in https://github.com/coqui-ai/TTS/pull/2783
• @wonkothesanest made their first contribution in https://github.com/coqui-ai/TTS/pull/2587
• @Subash-Lamichhane made their first contribution in https://github.com/coqui-ai/TTS/pull/3066
• @David-bfg made their first contribution in https://github.com/coqui-ai/TTS/pull/3027

Full Changelog: https://github.com/coqui-ai/TTS/compare/v0.17.8...v0.17.9

What's Changed

• XTTS redownload if needed by @Edresson in https://github.com/coqui-ai/TTS/pull/3038

Full Changelog: https://github.com/coqui-ai/TTS/compare/v0.17.7...v0.17.8