-
-
Notifications
You must be signed in to change notification settings - Fork 560
Pull requests: Blaizzy/mlx-audio
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
Improve performance of Parakeet TDT on longform content
#665
opened Apr 21, 2026 by
lucasnewman
Collaborator
Loading…
Fix Voxtral Realtime streaming and speed up the 4-bit path by ~3x
#661
opened Apr 20, 2026 by
iris-sfg
Loading…
2 of 3 tasks
feat (stt): Webserver: Diarization, export, Voxtral 3B Mini, larger audio files and enhanced model/language select
#643
opened Apr 10, 2026 by
Gotanius
Loading…
feat: add STS category to get_model_category and load_model
#642
opened Apr 8, 2026 by
TimPietrusky
Loading…
7 tasks done
feat: Add VoxCPM2 TTS model (2B params, 48kHz, 30 languages)
#641
opened Apr 8, 2026 by
acul3
Loading…
8 tasks done
Fix Voxtral TTS tokenizer dependency contract
#633
opened Apr 5, 2026 by
lyonsno
Loading…
3 tasks done
docs: add contributing guide, model porting guide, and code of conduct
#632
opened Apr 4, 2026 by
beshkenadze
Contributor
•
Draft
perf(fish_speech): lazy RAS sampling - only resample on repetition
#600
opened Mar 24, 2026 by
strongbinarycoder
Loading…
4 tasks done
fix: serialize MLX inference to prevent concurrent Metal command buffer crash
#594
opened Mar 22, 2026 by
hanai
Loading…
1 of 3 tasks
fix(parakeet): clamp multilingual v3 chunk windows for non-English input
stale
#589
opened Mar 20, 2026 by
nachoal
Loading…
1 of 3 tasks
Enable ref_audio and ref_text params in batch_generate and _prepare_batch_inputs for Qwen3-TTS model
#560
opened Mar 10, 2026 by
chartjin15
Loading…
3 tasks
Optimize ACE-Step: NLC VAE, compiled decode, LoRA support
#498
opened Feb 15, 2026 by
fspecii
Loading…
6 of 9 tasks
Fix VibeVoice-ASR streaming via generate() and server mapping
#483
opened Feb 5, 2026 by
JM27616
Loading…
1 of 3 tasks
Previous Next
ProTip!
Type g p on any issue or pull request to go back to the pull request listing page.