-
-
Notifications
You must be signed in to change notification settings - Fork 708
Pull requests: Blaizzy/mlx-audio
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
VibeVoice-ASR: keep speech features in the text embedding dtype (4-5x faster decoding)
#951
opened Sep 7, 2026 by
Oumnya
Loading…
fix(whisper): ignore empty non-speech token encodings
#934
opened Aug 30, 2026 by
saichowdary007
Loading…
fix(stt): stop fabricating Qwen3-ASR stream timestamps
#930
opened Aug 29, 2026 by
ainergiz
Contributor
Loading…
Release Please for faster release proceess
#922
opened Aug 28, 2026 by
Lazarus-931
Collaborator
Loading…
fix(server): bound resident model memory with LRU eviction and idle TTL sweeper
#913
opened Aug 27, 2026 by
rkfshakti
Loading…
3 tasks done
Onboarding: offline Dia tests + guide updates for quant predicates, KV cache sizing, and batch contracts
#909
opened Aug 24, 2026 by
mm65x
Contributor
Loading…
fix(qwen3_tts): yield silent audio instead of dropping empty batch items
#907
opened Aug 24, 2026 by
mm65x
Contributor
Loading…
docs(qwen3_tts): clarify voice cloning vs speaker synthesis usage rules
#906
opened Aug 24, 2026 by
mm65x
Contributor
Loading…
fix(fish_speech): prevent fast_ path and embeddings from quantizing to fix convert script
#905
opened Aug 24, 2026 by
mm65x
Contributor
Loading…
fix(dia): allocate kv cache to fit prefill + generation length
#904
opened Aug 24, 2026 by
mm65x
Contributor
Loading…
fix(server): stream one container per response instead of one per chunk
#899
opened Aug 21, 2026 by
arvindvenkataramani
Loading…
2 of 3 tasks
server: bound ModelProvider's model cache with LRU eviction
#884
opened Aug 11, 2026 by
mstkyvz
Loading…
feat: add CosyVoice3 MLX implementation
#861
opened Jul 28, 2026 by
walkoncross
Loading…
4 tasks done
Add CampPlus speaker-embedding conditioning to Ming Omni TTS (bailingmm)
#837
opened Jul 14, 2026 by
MukaiTensho
Loading…
Whisper: fall back to canonical openai/whisper-* processor when mlx-community repos lack one
#712
opened May 6, 2026 by
contrapuntal
Contributor
•
Draft
feat (stt): Webserver: Diarization, export, Voxtral 3B Mini, larger audio files and enhanced model/language select
#643
opened Apr 10, 2026 by
Gotanius
Loading…
docs: add contributing guide, model porting guide, and code of conduct
#632
opened Apr 4, 2026 by
beshkenadze
Contributor
•
Draft
Optimize ACE-Step: NLC VAE, compiled decode, LoRA support
#498
opened Feb 15, 2026 by
fspecii
Loading…
6 of 9 tasks
Previous Next
ProTip!
Follow long discussions with comments:>50.