v0.0.15

🐸 v0.0.15

🐞Bug Fixes

• Fix tb_logger init for rank > 0 processes in distributed training.

💾 Code updates

• Refactoring and optimization in the speaker encoder module. (:crown: @Edresson )
• Replacing unidecode with anyascii
• Japanese text to phoneme conversion. (:crown: @kaiidams)
• Japanese tts recipe to train Tacotron2-DDC on Kokoro dataset (:crown: @kaiidams)

🚶‍♀️ Operational Updates

• Start using pylint == 2.8.3
• Reorg tests files.
• Upload to pypi automatically on release.
• Move VERSION file under TTS folder.

🏅 Model implementations

• New Speaker Encoder implementation based on https://arxiv.org/abs/2009.14153 (:crown: @Edresson )

🚀 New Pre-Trained Model Releases

• Japanese Tacotron model (:crown: @kaiidams)

💡 All the models below are available by tts or tts-server endpoints on CLI as explained here.