Releases: Blaizzy/mlx-audio
Releases · Blaizzy/mlx-audio
v0.0.2
What's Changed
- fix workflows and readme by @Blaizzy in #5
- Add soundfile to requirements and Quick Start in README by @ivanfioravanti in #8
- Remove librosa dependency by @lucasnewman in #11
- Add support for command-line playback with the --play argument by @lucasnewman in #10
- Allow receiving text input from stdin or an entry prompt by @lucasnewman in #12
- Add web server and improve audio player by @ivanfioravanti in #14
- Use phonemizer-fork to avoid espeak errors by @rampadc in #17
New Contributors
- @Blaizzy made their first contribution in #5
- @ivanfioravanti made their first contribution in #8
- @lucasnewman made their first contribution in #11
- @rampadc made their first contribution in #17
Full Changelog: v0.0.1...v0.0.2
Contributors
ivanfioravanti, rampadc, and 2 other contributors
Assets 2
v0.0.1
Release Notes - February 28, 2025
Overview
This release introduces a fully functional MLX-Audio package with text-to-speech capabilities, complete with testing infrastructure and CI/CD integration via GitHub Actions.
New Features
- Text-to-Speech Generation: Added complete generation pipeline with audio output functionality
- Audio Joining: New functionality to join multiple audio segments
- Model Quantization: Added support for model quantization to improve performance
- GitHub Actions: Implemented CI/CD workflows for automated testing and deployment
Improvements
- Kokoro MLX porting: Completed refactoring of the entire model to MLX framework:
- Text encoder with BERT implementation
- Decoder with improved audio quality
- Duration, indices, and alignment target prediction
- Custom Bidirectional LSTM, Weight norm for CNNs, AdaLayerNorm and Generator layers
- SafeTensors Support: Added working implementation for SafeTensors format
- Pipeline Structure: Restructured the generation pipeline for better maintainability
Bug Fixes
- Fixed model loading mechanism
- Resolved issues with text encoder LayerNorm operation
- Fixed generator functionality
- Addressed issues in LSTM and AdaLayerNorm implementations
- Refactored and fixed ConvWeight component
Full Changelog: https://github.com/Blaizzy/mlx-audio/commits/v0.0.1