Commit Graph

Select branches

Hide Pull Requests

ScriptProcessorNode-to-AudioWorklet

VAD-evolutions

api_live

benchmarks-h100

feature/voxtral

language-detection

main

new-api

new_api

qwen3

regularfry-ayo-logging-fixes

rework_state_light

seamless-streaming

translation-improvements

voxtral_tests

windows_audio_loopback

#10

#103

#104

#108

#109

#113

#114

#116

#117

#118

#128

#145

#147

#148

#15

#153

#155

#161

#165

#166

#166

#170

#170

#172

#172

#173

#177

#178

#179

#183

#190

#190

#196

#198

#20

#200

#201

#211

#216

#217

#217

#219

#22

#221

#222

#223

#225

#233

#233

#237

#238

#239

#24

#249

#250

#259

#26

#260

#260

#267

#27

#272

#272

#273

#274

#285

#290

#297

#298

#298

#30

#303

#308

#314

#317

#322

#339

#34

#340

#341

#342

#35

#353

#355

#36

#4

#43

#45

#46

#47

#51

#52

#53

#55

#56

#57

#67

#68

#70

#71

#75

#76

#80

#83

#85

#91

#92

0.0.1

0.1.1

0.1.3

0.1.4

0.1.5

0.1.6

0.1.7

0.1.8

0.1.9

0.2.1

0.2.10

0.2.11

0.2.12

0.2.13

0.2.14

0.2.15

0.2.16

0.2.17

0.2.17.post1

0.2.18

0.2.2

0.2.4.dev0

0.2.5

0.2.6

0.2.7

0.2.8

0.2.9

v0.2.19

v0.2.20

Mono Color

• adaefeb26d Merge pull request #353 from brycehenson/fixes_cpu_demo main Quentin Fuxa 2026-05-13 10:27:49 +02:00
• a447065a99 Revert "change cuda" bryce henson 2026-04-07 22:13:41 +10:00
• d57e5a01f4 change cuda bryce henson 2026-04-07 22:12:37 +10:00
• 434b51befc another typo bryce henson 2026-04-03 02:02:56 +10:00
• ce1aed9632 typo bryce henson 2026-04-03 01:57:33 +10:00
• d505e72410 move ca-certificates to the later apt stage bryce henson 2026-04-03 01:56:18 +10:00
• 81fdba44ef remove remaining line from debugging in dockerfile bryce henson 2026-04-03 01:33:23 +10:00
• 26444efd3a fix certificates problem and missing python-multipart bryce henson 2026-04-03 01:28:59 +10:00
• 7fc341683a Remove unused benchmark and cascade bridge scripts Quentin Fuxa 2026-03-31 23:10:44 +02:00
• 3f233dc36c Fix all ruff lint errors (68 errors → 0) Quentin Fuxa 2026-03-31 23:02:50 +02:00
• db526ded34 Fix diarization failing on clips longer than ~1min (#349) Quentin Fuxa 2026-03-31 22:55:34 +02:00
• 3e5d8c5820 Fix Qwen3 streaming decode budget and head loading Quentin Fuxa 2026-03-23 23:03:11 +01:00
• b102e12943 M5 benchmark figures: WER vs RTF scatter, 0.6B+1.7B MLX results qwen3 Quentin Fuxa 2026-03-15 15:00:00 +01:00
• 7aa3b764bd MLX benchmark: 1.7B SimulStreaming on M5 (WER 4.07%, RTF 0.944) Quentin Fuxa 2026-03-15 14:00:00 +01:00
• a422e604ae MLX benchmark: 0.6B SimulStreaming on M5 MacBook (WER 3.30%, RTF 0.263) Quentin Fuxa 2026-03-15 13:00:00 +01:00
• e14b913807 Merge branch 'benchmarks-h100' Quentin Fuxa 2026-03-15 12:00:00 +01:00
• 47d4cbeecc reorganize benchmarks: move H100 results to benchmarks/h100/ benchmarks-h100 Quentin Fuxa 2026-03-15 23:59:00 +01:00
• 3b7a2fcc87 Add Qwen3-ASR MLX SimulStreaming backend Quentin Fuxa 2026-03-15 11:00:00 +01:00
• f75dfb386d final benchmark: Voxtral vLLM realtime streaming Quentin Fuxa 2026-03-15 23:59:00 +01:00
• 276ba84d02 update figures with Voxtral vLLM results Quentin Fuxa 2026-03-15 23:55:00 +01:00
• 36b3885cf2 add Voxtral 4B to benchmark figures Quentin Fuxa 2026-03-15 23:30:00 +01:00
• a29e799ba5 update H100 benchmark figures with ACL6060 results Quentin Fuxa 2026-03-15 22:30:00 +01:00
• 22325ba326 tune simul-kv: 2s inference interval, configurable min_new_seconds Quentin Fuxa 2026-03-15 21:30:00 +01:00
• a540a5fd10 fix simul-kv audio trim bug, add 1.7B v2 alignment heads Quentin Fuxa 2026-03-15 20:45:00 +01:00
• 7b08ea74ab add H100 benchmark figures Quentin Fuxa 2026-03-15 19:15:00 +01:00
• b69eaf82be qwen3 simul+kv: optimized streaming with kv cache reuse Quentin Fuxa 2026-03-15 18:30:00 +01:00
• 7ea507ed8e Add Voxtral MLX streaming backend feature/voxtral Quentin Fuxa 2026-02-17 09:20:28 +01:00
• ed503be140 qwen Quentin Fuxa 2026-01-02 23:52:00 +01:00
• a6a85431f6 update benchmark with qwen3 which reuses kv cache Quentin Fuxa 2026-03-15 22:32:01 +01:00
• dd48997674 qwen3: reuse encoder kv cache Quentin Fuxa 2026-03-15 22:31:39 +01:00
• f24481dc29 update archi Quentin Fuxa 2026-03-15 11:36:45 +01:00
• ed76f40ee5 Merge branch 'main' of https://github.com/QuentinFuxa/WhisperLiveKit Quentin Fuxa 2026-03-15 11:16:38 +01:00
• 5330b3fac5 update benchmark part Quentin Fuxa 2026-03-15 11:16:26 +01:00
• 0c73a73aa3 update benchmark results and procedure Quentin Fuxa 2026-03-15 11:16:15 +01:00
• 2d6bc4f572 Add '*.c' to .dockerignore Quentin Fuxa 2026-03-14 00:18:10 +01:00
• dfd5bf417c voxtral mlx : improved chunking Quentin Fuxa 2026-03-14 00:13:29 +01:00
• 9d8db7ab38 add qwen3 simul in tests Quentin Fuxa 2026-03-14 00:13:09 +01:00
• fa15115163 qwen3 alignment heads Quentin Fuxa 2026-03-14 00:12:50 +01:00
• 8dc7b77071 Bump version to 0.2.20 v0.2.20 Quentin Fuxa 2026-03-08 16:02:00 +01:00
• 10d85ff65f Update docs, CI, and architecture diagram Quentin Fuxa 2026-03-08 15:14:00 +01:00
• e7e3441ca4 Add Qwen3 ASR backend Quentin Fuxa 2026-03-07 11:48:00 +01:00
• 9abe26a996 Add CLI with serve, transcribe, listen, pull, diagnose Quentin Fuxa 2026-03-01 13:37:00 +01:00
• c8e7c216ed Replace mock tests with real pipeline tests Quentin Fuxa 2026-02-28 10:05:00 +01:00
• 586540ae36 Add test harness and test client Quentin Fuxa 2026-02-22 16:19:00 +01:00
• cd8df8e1aa Update package setup and exports Quentin Fuxa 2026-02-21 11:33:00 +01:00
• e30f9a2573 Improve diarization backends Quentin Fuxa 2026-02-15 14:55:00 +01:00
• 32de7b1276 Fix frontend buffer rendering for slow backends Quentin Fuxa 2026-02-14 09:28:00 +01:00
• 9ac7c26a0b Add OpenAI REST API and Deepgram WebSocket Quentin Fuxa 2026-02-08 15:42:00 +01:00
• c0e2600993 Add snapshot-then-diff WebSocket protocol Quentin Fuxa 2026-02-07 10:17:00 +01:00
• e0db3a98f9 Add per-session language proxy Quentin Fuxa 2026-02-01 17:03:00 +01:00
• 2fe34427ef Fix voxtral streaming drain and silence flush Quentin Fuxa 2026-01-31 11:12:00 +01:00
• d58365421f Refactor audio processor async pipeline Quentin Fuxa 2026-01-25 13:48:00 +01:00
• a282cbe75f Improve tokens alignment and silence handling Quentin Fuxa 2026-01-24 10:55:00 +01:00
• 6e85c16614 Refactor TranscriptionEngine singleton Quentin Fuxa 2026-01-18 15:27:00 +01:00
• e1823dd99c Improve online ASR processor Quentin Fuxa 2026-01-17 09:35:00 +01:00
• e144abbbc7 Refactor timed objects and data structures Quentin Fuxa 2026-01-11 16:08:00 +01:00
• 83362c89c4 Clean up config and model paths Quentin Fuxa 2026-01-10 11:42:00 +01:00
• 74c4dc791d Lint scripts and tests Quentin Fuxa 2026-01-04 14:15:00 +01:00
• cf6c49f502 Ruff lint cleanup Quentin Fuxa 2026-01-03 10:23:00 +01:00
• 451535d48f Fix ctranslate2 encoder conversion (#345) and memory leak in TokensAlignment (#344) Quentin Fuxa 2026-03-10 22:37:00 +01:00
• 8bc0937c46 Update README section on powered research Quentin Fuxa 2026-03-06 18:46:07 +01:00
• 929cf7a26b add link to AlignAtt interactive playground Quentin Fuxa 2026-03-06 18:43:25 +01:00
• abfaf06203 Merge branch 'main' of https://github.com/QuentinFuxa/WhisperLiveKit Quentin Fuxa 2026-03-04 18:17:23 +01:00
• d1fe932241 Apply DRY method v0 - to try to catch and resolve infinite loops such as in #338 Quentin Fuxa 2026-03-03 22:52:00 +01:00
• c112ceffb6 Merge pull request #342 from mnicnc404/fix/whisper-tokenizer-index-error Quentin Fuxa 2026-03-02 20:36:58 +01:00
• 4917406e06 Merge pull request #341 from AymurAI/feat/uv-deps-resolution Quentin Fuxa 2026-03-02 20:34:49 +01:00
• b63f54e838 fix(whisper/tokenizer): prevent IndexError from crashing multilingual streams Chingning Chen 2026-03-02 15:31:43 +08:00
• c56a53fbf4 deps(mlx-groups): add optional dependencies for Apple Silicon MLX backends jedzill4 2026-03-01 20:05:52 -03:00
• 66e58624b9 disable MLXAlignAtt which fails on special characters Quentin Fuxa 2026-03-01 11:52:00 +01:00
• 9366e067f9 deps(pyproject): add torch and torchaudio to main dependencies jedzill4 2026-02-27 19:19:18 -03:00
• 866c25670c deps(docker): change CUDA base image to runtime version jedzill4 2026-02-27 19:16:29 -03:00
• 2553ef283e deps(docker): fix dependency group for cu129 image jedzill4 2026-02-25 21:49:08 -03:00
• 73e7fafc48 feat(tests): python matrix support test jedzill4 2026-02-25 21:35:41 -03:00
• bbcebcb1fe deps(sortformer): adjust nemo-toolkit version constraints jedzill4 2026-02-25 21:33:00 -03:00
• 4bb58dc7aa deps(diart): improve diart dependency tree. rename gpu-cu129 dependency group to cu129 jedzill4 2026-02-25 20:27:26 -03:00
• 27ca028479 ci(github): add GitHub Actions workflows for Docker image publishing and support matrix jedzill4 2026-02-25 14:27:51 -03:00
• d24805cc18 🚀 chore (docker): update docker images improving caching and using uv as python package manager jedzill4 2026-02-25 14:22:43 -03:00
• 994ce21365 📌 chore(deps): pin dependences to python 3.11 to 3.13 due dependency resolution matrix jedzill4 2026-02-25 14:21:19 -03:00
• 132823dc09 deps: improve deps dependency resolution (wip) jedzill4 2026-02-24 20:15:53 -03:00
• d6d8c2635f chore: use uv as python project manager to improve dependency resolution jedzill4 2026-02-23 22:16:32 -03:00
• 8fedeb9fed Merge pull request #340 from QuentinFuxa/voxtral_tests v0.2.19 Quentin Fuxa 2026-02-23 10:37:40 +01:00
• b1fc23807a docs: add benchmark collaboration call, voxtral in powered-by section voxtral_tests Quentin Fuxa 2026-02-23 10:37:22 +01:00
• 10c4e5f730 docs: add speed vs accuracy scatter plot to benchmark and README Quentin Fuxa 2026-02-23 10:27:53 +01:00
• c76b2ef2c6 docs: rewrite benchmark with base/small comparison, proper French results Quentin Fuxa 2026-02-23 10:16:34 +01:00
• 4b2377c243 fix: correct false auto-detect claim, median bug, RTF inflation Quentin Fuxa 2026-02-22 23:38:04 +01:00
• a4da246ea5 feat: add voxtral-mlx native backend for Apple Silicon Quentin Fuxa 2026-02-22 23:28:10 +01:00
• 9b2c3ee844 docs: update README with voxtral backend, benchmarks, testing sections Quentin Fuxa 2026-02-22 23:27:57 +01:00
• 83d0fa3fac feat: benchmark suite with WER, timestamp accuracy, cross-backend comparison Quentin Fuxa 2026-02-22 23:27:50 +01:00
• 5a12c627b4 feat: add 99-test unit test suite with zero model dependencies Quentin Fuxa 2026-02-22 23:27:40 +01:00
• f5eee67b11 fix: silence double-counting bug, add metrics module and runtime instrumentation Quentin Fuxa 2026-02-22 23:27:12 +01:00
• 4a6868e3e1 correct processor attributes mixtral Quentin Fuxa 2026-02-22 21:13:21 +01:00
• 3c15246fc0 mixstral hf v0 Quentin Fuxa 2026-02-20 20:46:37 +01:00
• d337248fda feat: add healthcheck to Dockerfiles (#228) Quentin Fuxa 2026-02-19 22:18:00 +01:00
• b8d9d7d289 fix: handle numpy object_ dtype from ctranslate2 encoder (#337) Quentin Fuxa 2026-02-19 22:18:00 +01:00
• 4c7706e2cf fix: use vac_chunk_size for audio processing interval when VAC is enabled (#334) Quentin Fuxa 2026-02-19 22:18:00 +01:00
• d9a4c8dcb2 Refactor transcription and diarization handling with token-by-token validation. Introduce segment buffers for ephemeral content and update API to return structured segment data. Enhance silence handling and improve web interface for text transcripts. api_live Quentin Fuxa 2025-11-30 16:39:27 +01:00
• 4fb735a784 new token treatment only iar Quentin Fuxa 2025-11-30 15:16:36 +01:00
• d2f998cb7e val Quentin Fuxa 2025-11-30 14:37:37 +01:00
• 7b18917f2b LoRA archi Quentin Fuxa 2025-11-30 12:30:18 +01:00
• 9d4ae33249 WIP. Trying ten VAD #280 VAD-evolutions Quentin Fuxa 2025-11-23 11:20:00 +01:00