| Name | Modified | Size | Downloads / Week |
|---|---|---|---|
| Parent folder | |||
| README.md | 2026-08-10 | 1.2 kB | |
| v0.4.8 source code.tar.gz | 2026-08-10 | 6.8 MB | |
| v0.4.8 source code.zip | 2026-08-10 | 7.2 MB | |
| Totals: 3 Items | 14.0 MB | 0 | |
What's Changed
- Use the high quality resampler for miniaudio-loaded audio files by @lucasnewman in https://github.com/Blaizzy/mlx-audio/pull/872
- Ignore unknown decode options in Whisper generate by @Lazarus-931 in https://github.com/Blaizzy/mlx-audio/pull/873
- fix(kokoro, kitten_tts): match PyTorch reference semantics in five decoder paths (-2.5 dB level error, F0-path misalignment, window mismatch) by @mchen04 in https://github.com/Blaizzy/mlx-audio/pull/859
- Add Nemotron VoiceChat codec by @lucasnewman in https://github.com/Blaizzy/mlx-audio/pull/876
- Fix unintentional float32 upcast for Qwen3-TTS by @lucasnewman in https://github.com/Blaizzy/mlx-audio/pull/879
- Improved Nemotron ASR streaming by @lucasnewman in https://github.com/Blaizzy/mlx-audio/pull/878
- Fix model type resolution for Higgs Audio v3 by @lucasnewman in https://github.com/Blaizzy/mlx-audio/pull/868
- Bump version to 0.4.8 by @Blaizzy in https://github.com/Blaizzy/mlx-audio/pull/883
New Contributors
- @mchen04 made their first contribution in https://github.com/Blaizzy/mlx-audio/pull/859
Full Changelog: https://github.com/Blaizzy/mlx-audio/compare/v0.4.7...v0.4.8