modelscope/FunASR
1
0
Fork 0
mirror of https://github.com/modelscope/FunASR.git synced 2026-10-12 02:19:18 +00:00
Code Issues Packages Projects Releases Wiki Activity
5,117 Commits 468 Branches 66 Tags
2be876f437e48ae01b03aa978c03c4faf187d745
Commit Graph
388 Commits
Author SHA1 Message Date
游雁 5a37a31462 docs: avoid unbounded vLLM install guidance 2026-06-11 15:54:58 +00:00
游雁 ef46055056 fix: respect explicit VAD silence threshold 2026-06-11 15:16:24 +00:00
游雁 b78e636170 docs: add CLI page to sphinx toctree 2026-06-10 09:11:29 +00:00
游雁 cffa9f2c55 docs: add CLI reference documentation 2026-06-10 09:10:05 +00:00
游雁 1065d2e5ab revert: remove incorrect VAD finetune docs and script 
FsmnVADStreaming.forward() is inference-only (no loss computation),
so standard train_ds.py training does not work. Removing the
finetune.sh and docs that were incorrectly added.

The parameter tuning guide (vad_kwargs) remains valid and is
documented in the issue thread (#2961).
 2026-06-10 03:46:57 +00:00
游雁 80d49eefaf chore: remove tracked .DS_Store file 2026-06-09 12:49:30 +00:00
游雁 e7ba9b3c3e docs: update Python version support to 3.8-3.13 2026-06-08 11:17:33 +00:00
游雁 ef0e46f1c3 docs: add VAD fine-tuning guide and finetune script 
- Add finetune.sh for FSMN-VAD model fine-tuning
- Add English and Chinese documentation for VAD fine-tuning
- Include parameter tuning guide for quick adjustments without fine-tuning
- Addresses user requests in #2961
 2026-06-08 11:14:47 +00:00
游雁 21e6ededcc docs: add Chinese blog draft — FunASR vs Whisper comparison for Zhihu/CSDN 2026-05-28 08:14:57 +08:00
zhifu gao 5d8349aeac docs: add torch+torchaudio to vllm_guide_zh install section 2026-05-27 21:17:38 +08:00
zhifu gao 768750587d docs: add torch+torchaudio to vllm_guide install section 2026-05-27 21:17:20 +08:00
zhifu gao 31ac626c89 docs: add Japanese and Korean deployment matrix guides 2026-05-26 17:47:46 +08:00
zhifu gao e29b4a999a docs: add Japanese and Korean model selection guides 2026-05-26 17:02:57 +08:00
游雁 8325d5eabf docs: rewrite English vLLM guide with high-quality manual translation 2026-05-26 12:08:07 +08:00
游雁 0f780ce592 docs: rename to '离线 SDK 推理' + add design principles, sync zh/en/v2 2026-05-26 12:05:07 +08:00
游雁 21e0889b1c docs: rename to '离线 SDK 批量推理' and add design principles section 2026-05-26 11:55:30 +08:00
游雁 8ca7fb2b8c docs: add vllm_guide_zh_v2.md — merged batch inference + streaming SDK into service guide 2026-05-26 02:39:52 +08:00
游雁 bb92a779fd docs: restore vLLM guide — add English version and Chinese (vllm_guide_zh.md) 
The original docs/vllm_guide.md was accidentally emptied. Restored the full
content as docs/vllm_guide_zh.md (Chinese) and created docs/vllm_guide.md
(English translation) as the primary guide.
 2026-05-26 02:25:58 +08:00
游雁 7d3c2161db feat: WebSocket SPK as client-controlled param (SPK:true/false, default off) 2026-05-26 02:12:31 +08:00
游雁 a67e7d7356 feat: add SPK support to all protocols (OpenAI API + WebSocket) 
- OpenAI API: added spk param (FunASR extension)
- WebSocket: SPK clustering on STOP (full re-cluster all sentences)
- All 3 protocols tested: HTTP REST ✓, OpenAI API ✓, WebSocket ✓
- docs: merge VAD tables into one, remove "no SPK" notes
 2026-05-26 02:07:40 +08:00
游雁 374afebb39 docs: note WebSocket also no SPK support 2026-05-26 02:00:52 +08:00
游雁 800bd2cd30 docs: fix offline service - split by protocol with per-protocol clients, add WebSocket, note OpenAI no SPK 2026-05-26 01:57:56 +08:00
游雁 ba2a3ac4f0 docs: final vllm_guide - benchmark first, architecture section, clear service separation 2026-05-26 01:47:06 +08:00
游雁 4cc2b75446 docs: rewrite vllm_guide - separate offline/streaming services, full protocol + client examples 2026-05-26 01:40:10 +08:00
游雁 e46abebc88 docs: rewrite vllm_guide.md - unified structure, dynamic VAD only, stream/offline separation 2026-05-26 01:33:57 +08:00
游雁 6a18244602 feat: finalize vLLM service + benchmark matrix 
Updates:
- Non-streaming dynamic VAD: (10000,2000)→(60000,200)→(inf,100)
  Fun-ASR-Nano CER 8.90%→8.20%; GLM-ASR not supported (long audio)
- SPK is opt-in (default off), configurable via API param spk=true
- serve_vllm.py: unified HTTP/OpenAI/WebSocket server
- benchmark_vllm.py: unified benchmark for all models
- docs/vllm_guide.md: full benchmark matrix (8 configs)

Benchmark (184 files, 11541s):
  Fun-ASR-Nano vLLM batch (fixed): RTFx=371, CER=8.88%
  Fun-ASR-Nano vLLM batch (dynamic): RTFx=340, CER=8.20%
  Fun-ASR-Nano Service (no SPK): RTFx=102, CER=8.14%
  Fun-ASR-Nano Service (with SPK): RTFx=46, CER=8.19%
  GLM-ASR vLLM (fixed): RTFx=265, CER=12.93%
 2026-05-26 01:26:09 +08:00
游雁 005e9d8458 merge vllm: new dynamic VAD + unified server + updated benchmark 2026-05-25 23:22:45 +08:00
游雁 3bc9018854 feat: add unified vLLM inference server (serve_vllm.py) 
Three interfaces:
- HTTP REST: POST /asr (file upload, VAD + ASR + SPK + timestamps)
- OpenAI API: POST /v1/audio/transcriptions (Whisper-compatible)
- WebSocket: ws://host:port/ws (streaming with dynamic VAD)

All share: vLLM engine + dynamic VAD + speaker diarization + CTC timestamps.
Performance: 227s audio processed in 3.4s (RTF=0.015), CER=9.30%.
 2026-05-25 22:20:14 +08:00
游雁 6a89af3847 docs: add CTC timestamp output example to vllm_guide.md 2026-05-25 22:03:41 +08:00
游雁 dd72f933be Merge branch 'main' of https://github.com/modelscope/FunASR 2026-05-25 20:29:24 +08:00
游雁 06558dfdd2 docs: merge benchmark tables into single unified table 2026-05-25 20:29:12 +08:00
zhifu gao fc76d9945b docs: localize Colab quickstart guide 2026-05-25 20:27:52 +08:00
游雁 653d787c21 refactor: unify benchmark_vllm.py at project root 
- Single script supports all models (Fun-ASR-Nano, GLM-ASR, AutoModelVLLM)
- Remove old model-specific benchmark from examples/
- Update docs with unified benchmark commands
 2026-05-25 20:22:12 +08:00
游雁 674ba5e53e docs: add GLM-ASR vLLM benchmark results to vllm_guide.md 2026-05-25 20:09:24 +08:00
zhifu gao f31c1c4f8a docs: add Colab quickstart notebook 
Add a browser-run FunASR Colab quickstart and link it from README, tutorial, use-case, and deployment docs.
 2026-05-25 19:40:27 +08:00
游雁 1ec9c0ab69 Merge branch 'main' of https://github.com/modelscope/FunASR 2026-05-25 19:28:25 +08:00
游雁 4484557424 docs: update vllm_guide.md with benchmark results (RTFx=393.9, CER=8.91%) 2026-05-25 19:28:13 +08:00
zhifu gao eef89baec8 docs: add OpenAI API security gateway guide 
Add English and Chinese gateway hardening docs for the OpenAI-compatible API and link them from deployment and recipe docs.
 2026-05-25 19:13:36 +08:00
zhifu gao 6165c19990 demo: add Gradio OpenAI API browser demo 
Add an optional Gradio browser UI and docs for the FunASR OpenAI-compatible API.
 2026-05-25 18:37:40 +08:00
zhifu gao 0d037a36ed docs: add model selection guide 
Add English and Chinese model selection guidance for FunASR models, OpenAI API aliases, and deployment paths.
 2026-05-25 18:19:28 +08:00
zhifu gao 8ade9bdb5e docs: add Kubernetes OpenAI API deployment template 
Add CPU-first Kubernetes manifests and deployment docs for the FunASR OpenAI-compatible API.
 2026-05-25 17:52:43 +08:00
zhifu gao 31119ec0e5 docs: add JavaScript OpenAI API recipes 
Document OpenAI JS SDK, Node fetch, TypeScript helper, and Next.js proxy patterns for FunASR OpenAI-compatible API users.
 2026-05-25 17:09:38 +08:00
zhifu gao 0386bff0a4 test: add Python OpenAI API smoke test 
Add a cross-platform Python smoke test and document it alongside the existing bash smoke test.
 2026-05-25 16:35:26 +08:00
zhifu gaoandxiaoyunchong.xyc 19c62e9713 docs: add Chinese Postman and OpenAPI guides 
Co-authored-by: xiaoyunchong.xyc <[email protected]>
 2026-05-25 15:57:01 +08:00
zhifu gaoandxiaoyunchong.xyc 9f494cbae7 docs: add Chinese OpenAI API quickstart 
Co-authored-by: xiaoyunchong.xyc <[email protected]>
 2026-05-25 15:28:42 +08:00
zhifu gaoandxiaoyunchong.xyc 46bc2c7759 docs: add OpenAI API OpenAPI spec 
Co-authored-by: xiaoyunchong.xyc <[email protected]>
 2026-05-25 14:58:01 +08:00
zhifu gaoandxiaoyunchong.xyc 2549bbc3bd docs: add Chinese workflow recipes 
Co-authored-by: xiaoyunchong.xyc <[email protected]>
 2026-05-25 13:28:20 +08:00
zhifu gao fa2e1a7f80 docs: add OpenAI API workflow recipes 
Add low-code workflow recipes for Dify, n8n, HTTP nodes, and webhook workers using the OpenAI-compatible FunASR API.
 2026-05-25 12:26:36 +08:00
zhifu gao d0da2adfd0 docs: add migration benchmark issue template 
Add a dedicated migration benchmark report template and link it from migration docs and examples.
 2026-05-25 11:26:35 +08:00
zhifu gao cbae12f1f1 docs: add migration benchmark example 
Add a runnable migration benchmark example for representative FunASR audio evaluation and link it from README/docs.
 2026-05-25 10:28:14 +08:00