Starstruk Logo MarkStarstruk

v3.1

nari-labs/dia2v3.1Jul 25, 2024by Flux9665

AI Summary

This release refines the universal TTS model by introducing stochastic prosody prediction and expanding language support capabilities. It serves as a refinement of the v3.0 release to address initial limitations and improve speech generation quality.

Key Highlights

  • Stochastic prosody prediction for pitch, energy, and durations
  • Expanded support for IPA modifiers to cover more languages
  • Increased languages included in the pretraining dataset
  • Overhauled language similarity prediction modules

New Features

  • Stochastic prosody prediction
  • Expanded IPA modifiers
  • More languages in pretraining
  • Overhauled language similarity prediction modules and visualization

Full Release Notes

## What's Changed

This release provides new checkpoints and improves some aspects of the previous release that were not included due to time constraints.  For more information on the universal TTS model for 7000 languages, please refer to the previous release v3.0

- Prosody prediction in terms of pitch, energy and durations are now stochastic and sample from a distribution instead of assuming a one-to-one mapping. 
- Added support for more IPA modifiers to cover more languages
- Added more languages into the pretraining 
- Overhauled language similarity prediction modules and visualization

**Full Changelog**: https://github.com/DigitalPhonetics/IMS-Toucan/compare/v3.0...v3.1