Commits

Branch selector

main

User selector

All users

Datepicker

All time

Commit history

Commits on Oct 7, 2026

• server: optional input on /v1/audio/speech/live for s2s and vc models (#822)

  Show description for a1c24b2
  ykhrustalev

  authored
  a1c24b2View commit details

  Copy full SHA for a1c24b2

  Browse repository at this point

• Reduce Qwen3-TTS, IndexTTS, and BreezeTTS memory allocations (#821)

  Show description for 6fa5a3c
  0xShug0

  authored
  6fa5a3cView commit details

  Copy full SHA for 6fa5a3c

  Browse repository at this point

Commits on Oct 6, 2026

• fix(lfm2_audio): refuse audio too long for one encoder pass, and name missing package files (#820)

  Show description for a08c7af
  ykhrustalev

  authored
  a08c7afView commit details

  Copy full SHA for a08c7af

  Browse repository at this point

• Relax Niagara ASR and AuK backend restrictions (#817)

  Show description for 0b1e613
  0xShug0

  authored
  0b1e613View commit details

  Copy full SHA for 0b1e613

  Browse repository at this point

• Sopro v2.2.0 parity, real (chunked) streaming, and opt-in voice cache (#818)

  Show description for d331c27
  samuel-vitorino

  authored
  d331c27View commit details

  Copy full SHA for d331c27

  Browse repository at this point

• docs(reports): add YuE2 longform host commit exhaustion report (#816)

  Show description for f31ab33
  wangguanghua2099

  authored
  f31ab33View commit details

  Copy full SHA for f31ab33

  Browse repository at this point

• refactor(lfm2_audio): choose the decode-cache size by task (#813)

  Show description for f78d6f3
  ykhrustalev

  authored
  f78d6f3View commit details

  Copy full SHA for f78d6f3

  Browse repository at this point

• Fix Stable Audio short-input encoder stitching (#815)

  0xShug0

  authored
  3878865View commit details

  Copy full SHA for 3878865

  Browse repository at this point

• Document core model options with compatible normalized names (#814)

  0xShug0

  authored
  19859a0View commit details

  Copy full SHA for 19859a0

  Browse repository at this point

• Fix ZipEnhancer graph constants and stable activations (#811)

  0xShug0

  authored
  d894789View commit details

  Copy full SHA for d894789

  Browse repository at this point

• Add BreezeTTS2 Q4 package and PocketTTS voice export docs (#808)

  0xShug0

  authored
  bfb13e2View commit details

  Copy full SHA for bfb13e2

  Browse repository at this point

• 🚨 Accept neutral speed in OpenAI speech requests (#807)

  Show description for bd79d7d
  0xShug0

  authored
  bd79d7dView commit details

  Copy full SHA for bd79d7d

  Browse repository at this point

• feat: return_text on /v1/audio/speech/live (#805)

  Show description for 6bb6add
  ykhrustalev

  authored
  6bb6addView commit details

  Copy full SHA for 6bb6add

  Browse repository at this point

• Use depthwise upsampling for PocketTTS on CPU (#806)

  0xShug0

  authored
  216e1c6View commit details

  Copy full SHA for 216e1c6

  Browse repository at this point

• docs: add ComeCut to the Projects list (#804)

  juntaosun

  authored
  5b35e90View commit details

  Copy full SHA for 5b35e90

  Browse repository at this point

Commits on Oct 5, 2026

• feat: add lfm2_audio speech-to-speech for LFM2.5-Audio EN and JP (#798)

  Show description for 8fa20ff
  ykhrustalev

  authored
  8fa20ffView commit details

  Copy full SHA for 8fa20ff

  Browse repository at this point

• fix: allow --out to target FIFOs and char devices (e.g. /dev/stdout) (#796)

  trappedinspacetime

  authored
  54aa279View commit details

  Copy full SHA for 54aa279

  Browse repository at this point

• Enhance CUDA 12 library handling in Colab notebook (#795)

  Show description for adf3782
  ha466

  authored
  adf3782View commit details

  Copy full SHA for adf3782

  Browse repository at this point

• Gate KittenTTS2 preparation tool behind optional test builds

  0xShug0

  committed
  3cf2161View commit details

  Copy full SHA for 3cf2161

  Browse repository at this point

• Fix repeated Voxtral Realtime transcriptions (#793)

  dignome

  authored
  50d6886View commit details

  Copy full SHA for 50d6886

  Browse repository at this point

Commits on Oct 4, 2026

• server: add include_params to GET /v1/models, read from the model spec (#787)

  Show description for ad9bc82
  

  Navier8

  and

  claude

  authored
  ad9bc82View commit details

  Copy full SHA for ad9bc82

  Browse repository at this point

• Document OmniVoice and Qwen3 options with compatible normalized names (#786)

  0xShug0

  authored
  c6b104eView commit details

  Copy full SHA for c6b104e

  Browse repository at this point

• Reduce HTDemucs inference overhead across backends (#790)

  Show description for 24b69c9
  0xShug0

  authored
  24b69c9View commit details

  Copy full SHA for 24b69c9

  Browse repository at this point

• Add tts-audiobook-tool to projects listing in README.md (#788)

  zeropointnine

  authored
  2721dc0View commit details

  Copy full SHA for 2721dc0

  Browse repository at this point

• Align UI catalog model paths with spec packages (#783)

  0xShug0

  authored
  d487a65View commit details

  Copy full SHA for d487a65

  Browse repository at this point

• feat: add lfm2_audio TTS for LFM2.5-Audio EN and JP (#768)

  Show description for 42faa8b
  ykhrustalev

  authored
  42faa8bView commit details

  Copy full SHA for 42faa8b

  Browse repository at this point

• Egyptian arabic voicetut qwencleo (#780)

  Show description for 1487b9f
  

  MohammedAly22

  and

  Mohammed Aly

  authored
  1487b9fView commit details

  Copy full SHA for 1487b9f

  Browse repository at this point

• Fix Kitten TTS 2 reported wall_ms and clean up model implementation (#781)

  Show description for 56117a7
  dignome

  authored
  56117a7View commit details

  Copy full SHA for 56117a7

  Browse repository at this point

• Fix WebUI Run for bundled VAD models (#782)

  Show description for d8a9e2d
  mister2d

  authored
  d8a9e2dView commit details

  Copy full SHA for d8a9e2d

  Browse repository at this point

• Add Kittentts2 community model (#776)

  Show description for d3ab9df
  dignome

  authored
  d3ab9dfView commit details

  Copy full SHA for d3ab9df

  Browse repository at this point

• Fix Chatterbox S3 relative shift and encoder LayerNorm epsilon (#778)

  dignome

  authored
  07da27dView commit details

  Copy full SHA for 07da27d

  Browse repository at this point

• server: separate legacy and experimental parallel jobs runtimes (#715)

  Show description for e3e1bc7
  mirek190

  authored
  e3e1bc7View commit details

  Copy full SHA for e3e1bc7

  Browse repository at this point

Commits on Oct 3, 2026

• Add OWSM v4 and OWSM-CTC v4 (#774)

  Show description for 2ba9fa8
  0xShug0

  authored
  2ba9fa8View commit details

  Copy full SHA for 2ba9fa8

  Browse repository at this point

• Use UTF-8 process code pages for Windows executables (#772)

  Show description for ba6f360
  

  0xShug0

  and

  pnnbao

  authored
  ba6f360View commit details

  Copy full SHA for ba6f360

  Browse repository at this point

• docs(r2t2): note that the final word needs trailing silence (#770)

  Show description for 93c3240
  

  scriptease

  and

  claude

  authored
  93c3240View commit details

  Copy full SHA for 93c3240

  Browse repository at this point