Skip to content

v2.1.0

Choose a tag to compare

@wildminder wildminder released this 14 Apr 22:13
· 40 commits to main since this release

ComfyUI-VoxCPM v2.1.0

Audio processing improvements and frontend enhancements.


▓ What's New

** Audio Processing Backend**

  • Replaced librosa with torchaudio as the default audio processing backend for improved performance and GPU compatibility
  • Added abstraction layer supporting multiple backends (torchaudio default, librosa fallback)
  • Reduced dependency footprint by removing librosa requirement

Frontend Notifications

  • Added real-time notifications for missing text normalization packages
  • Improved user feedback for configuration issues

Bug Fixes

  • Fixed text normalization fallback handling
  • Fixed ProgressBar integration for ComfyUI UI feedback

Full Changelog

v2.0.1...v2.1.0