{"full_name":"modelscope/FunASR","owner":"modelscope","name":"FunASR","shard":"1","url":"https://github.com/modelscope/FunASR","description":"Open-source speech recognition toolkit for training, inference, streaming ASR, VAD, punctuation, speaker diarization pipelines, and OpenAI-compatible/MCP serving.","homepage":"www.funasr.com","topics":["pytorch","speech-recognition","paraformer","punctuation","speaker-diarization","voice-activity-detection","asr","multilingual-asr","speech-to-text","transcription","whisper-alternative","audio","chinese","emotion-recognition","mcp-server"],"tag_pages":["asr","audio","chinese","mcp-server","pytorch","speech-recognition","speech-to-text","transcription"],"language":"Python","license":"MIT","avatar":"https://avatars.githubusercontent.com/u/109945100?v=4","created_at":"2022-11-24T02:28:11Z","pushed_at":"2026-11-15T17:33:49Z","metrics":{"stars":21033,"forks":2129,"watchers":113,"issues":1809,"prs":2001,"releases":87,"issues_open":44,"issues_closed":1765,"prs_open":11,"prs_closed":157,"prs_merged":1833,"discussions":66},"growth":{"day":{"stars":8,"forks":1,"base_day":"2026-11-15"},"week":{"stars":109,"forks":23,"base_day":"2026-11-04"},"month":{"stars":329,"forks":48,"base_day":"2026-10-17"}},"history":{"dates":["2026-09-06","2026-09-07","2026-09-08","2026-09-09","2026-09-10","2026-09-11","2026-09-12","2026-09-13","2026-09-14","2026-09-15","2026-09-16","2026-09-17","2026-09-18","2026-09-19","2026-09-20","2026-09-21","2026-09-22","2026-09-23","2026-09-24","2026-09-25","2026-09-26","2026-09-27","2026-09-28","2026-09-29","2026-09-30","2026-10-01","2026-10-02","2026-10-03","2026-10-04","2026-10-05","2026-10-06","2026-10-07","2026-10-08","2026-10-09","2026-10-10","2026-10-11","2026-10-12","2026-10-13","2026-10-14","2026-10-15","2026-10-16","2026-10-17","2026-10-18","2026-10-19","2026-10-20","2026-10-21","2026-10-22","2026-10-23","2026-10-24","2026-10-25","2026-10-26","2026-10-27","2026-10-28","2026-10-29","2026-10-30","2026-10-31","2026-11-01","2026-11-02","2026-11-03","2026-11-04","2026-11-10","2026-11-11","2026-11-12","2026-11-13","2026-11-14","2026-11-15","2026-11-16"],"stars":[20172,20185,20199,20219,20230,20248,20265,20282,20284,20306,20322,20348,20372,20393,20417,20431,20443,20458,20474,20481,20500,20506,20508,20518,20535,20549,20559,20568,20569,20576,20580,20586,20594,20602,20613,20622,20636,20645,20662,20677,20693,20704,20718,20745,20754,20768,20785,20795,20800,20804,20812,20827,20843,20861,20874,20887,20895,20902,20913,20924,20985,20998,21006,21008,21018,21025,21033],"forks":[2014,2014,2014,2014,2021,2022,2023,2025,2025,2029,2029,2030,2033,2036,2036,2039,2040,2043,2043,2043,2043,2045,2047,2048,2051,2051,2053,2054,2056,2057,2058,2059,2060,2064,2066,2067,2073,2075,2075,2078,2079,2081,2083,2085,2087,2091,2091,2091,2092,2093,2093,2096,2095,2095,2096,2099,2101,2103,2102,2106,2116,2122,2123,2124,2127,2128,2129],"watchers":[118,118,119,119,118,118,118,118,118,118,118,118,119,119,119,119,119,119,119,119,120,120,120,120,118,118,118,118,118,118,118,118,118,118,118,118,118,118,118,118,118,118,118,118,118,118,118,118,117,114,114,114,114,114,114,114,114,114,114,114,null,113,113,113,113,113,113],"issues_open":[null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,46,47,45,44,44,44],"prs_open":[null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,14,15,12,13,12,11]},"boards":{},"same_owner":[{"full_name":"modelscope/facechain","name":"facechain","shard":"1","boards":{}},{"full_name":"modelscope/modelscope","name":"modelscope","shard":"1","boards":{}},{"full_name":"modelscope/ms-swift","name":"ms-swift","shard":"1","boards":{}}],"issues":[{"number":3302,"title":"增加Hotwords會造成語句變短及亂取代其他文字","comments":34,"created_at":"2026-07-22T09:28:23Z"},{"number":3203,"title":"建议给一个 roadmap","comments":28,"created_at":"2026-07-12T14:24:56Z"},{"number":3479,"title":"[Bug] Vulkan backend crashes on AMD GPUs on Windows (0xc0000005)","comments":22,"created_at":"2026-08-07T12:04:36Z"},{"number":3038,"title":"关于serve_realtime_ws.py服务的客户端测试达不到实时","comments":16,"created_at":"2026-06-29T08:02:59Z"},{"number":3773,"title":"中文断句的老问题，才发现代码调用和命令行方式运行的结果好像是有差别的，不一定适用所有音频文件哈","comments":11,"created_at":"2026-10-14T02:20:31Z"},{"number":3825,"title":"Fun-ASR-Nano-2512、SenseVoice复合流水线的 NPU 性能/兼容性","comments":9,"created_at":"2026-11-02T03:41:43Z"},{"number":3496,"title":"Fun-ASR-Nano Hugging Face checkpoint is missing 86 CTC tensors required for timestamps and diarization","comments":9,"created_at":"2026-08-14T08:20:41Z"},{"number":3388,"title":"SenseVoice-Small 增量微调:保留普通话+粤语的前提下新增潮汕话识别,如何避免灾难性遗忘?","comments":9,"created_at":"2026-07-23T02:17:50Z"}],"discussions":[{"number":3900,"title":"License confirmation for bundling sherpa-onnx-paraformer-zh-small-2024-03-09 in a closed-source commercial Android app","comments":1,"category":"Q&A","created_at":"2026-11-14T07:02:58Z"},{"number":3720,"title":"FunASR v1.4.16: reliable mixed-device submodels and native Transformers guides","comments":0,"category":"Announcements","created_at":"2026-09-18T06:20:01Z"},{"number":3642,"title":"FunASR v1.4.14: portable source archives and verified runtimes","comments":0,"category":"Announcements","created_at":"2026-09-04T04:21:02Z"},{"number":3586,"title":"MOSS-Transcribe-Diarize joins the FunASR ecosystem / MOSS 一体化转写与说话人分段已接入 FunASR","comments":0,"category":"Announcements","created_at":"2026-08-30T20:57:36Z"},{"number":3536,"title":"FunASR 1.4.5: lighter Python install and nine-platform llama.cpp runtimes / 更轻安装与九平台运行时","comments":0,"category":"Announcements","created_at":"2026-08-27T17:55:27Z"},{"number":3525,"title":"Possible Paraformer timestamp issues in the Hugging Face full pipeline and GPU dynamic batching","comments":4,"category":"General","created_at":"2026-08-25T03:38:23Z"},{"number":3521,"title":"FunASR 1.4.3: optional Silero VAD and fixed-K speaker clustering / 可选 Silero VAD 与固定 K 聚类","comments":0,"category":"Announcements","created_at":"2026-08-21T04:15:21Z"},{"number":3513,"title":"FunASR-nano-2512 + vLLM的AutoModelVLLM是否可以支持输出sentence_info句子级别的时间戳","comments":1,"category":"Q&A","created_at":"2026-08-17T08:32:55Z"}],"latest_release":{"name":"v1.4.26","tag":"v1.4.26","published_at":"2026-11-12T19:42:51Z","body":"<!-- release1426-pypi-verified-20261112 -->\n### Final Publication Acceptance\n\nGitHub Release and PyPI 1.4.26 are published. Both official PyPI distributions were downloaded and match GitHub by SHA-256 and size:\n\n- `funasr-1.4.26-py3-none-any.whl`: 1015738 bytes; SHA-256 `104e3704fed0ecfd6c08836468de147a2c6f05745281093b8a6dbe341d1ce831`.\n- `funasr-1.4.26.tar.gz`: 1086172 bytes; SHA-256 `763c7d860a49afd17e4b9c0636ff43f03f6519e68bd0a09a25d68318091e2cf8`.\n\n<!-- release1426-final-verification-20261112 -->\n## FunASR 1.4.26\n\nSource: `cdd82b5cbfd05015f8fb2b167d61f25142c3a9e8` / tag `v1.4.26`.\n\n### Fixes\n\n- #3874: `response_format=text` now returns the bare HTTP transcript instead of a JSON-quoted string.\n- #3872 / #3883: retain word separators across English and Korean VAD segments while preserving CJK and event-only behavior. The original author's commit is retained.\n\nThe independent ONNX/Torch abbreviation and batch-padding timestamp fixes are not shipped by this root wheel. Their package releases are separate. Open #3861, #3862, #3867 and the new hotword PRs are not claimed resolved.\n\n### Verification Scope\n\n- #3874 previously passed 32 actual HTTP/SDK cases and 213 related CPU cases; its source bytes remain unchanged here.\n- #3883 passed 246 unique selected CPU cases across the initial 230 passes and completion of six optional-dependency cases plus ten actual installed standalone-wheel cases. All original skipped and negative runs remain preserved. NumPy hosted CI has 421 passes and 27 CUDA-only skips per version, not GPU acceptance.\n- Release/docs candidate checks: {'contract-green': {'testcase': 53, 'failure': 0, 'error': 0, 'skipped': 0}, 'static-remainder': {'testcase': 104, 'failure': 0, 'error': 0, 'skipped': 2}, 'installed-behavior-candidate': {'testcase': 111, 'failure': 0, 'error': 0, 'skipped': 0}, 'site-boundary-integration': {'testcase': 69, 'failure': 0, 'error': 0, 'skipped': 0}, 'site-manifest-integration': {'testcase': 21, 'failure': 0, 'error': 0, 'skipped': 0}}. Actual installed root wheel final counts: {'testcase': 111, 'failure': 0, 'error': 0, 'skipped': 0}. Standalone runtime paths are explicitly deselected from root-wheel tests. All 439 package files and 13 entrypoints match accepted source.\n- Root model/GPU-runtime bytes are unchanged from 1.4.25 and remain bound to its accepted prior L20Z evidence. No new GPU execution, pretrained ASR-quality, full-suite or Nano result is claimed.\n- Head, merged-main and final-tag hosted CI and actual checkout trees are separately accepted. Final Python wheel and sdist are newly downloaded and checked for metadata, exact source bytes, archive integrity, strict Twine, private offline installation and CLI behavior.\n- Ten unchanged runtime archives reuse complete cache bytes, freshly rehashed against the new assets and original runtime release with archive-integrity checks. They were not rebuilt or re-executed. SHA256SUMS covers all 12 payloads.\n- The 1.4.24 warning still applies: checkpoints trained with both norm_means=True and norm_vars=True need recognition-quality reevaluation. norm_vars=False remains the default.\n- No production upgrade or adoption/growth claim.\n\n## What's Changed\n* fix(runtime): release ONNX non-WAV audio loading as 0.4.10 by @LauraGPT in https://github.com/modelscope/FunASR/pull/3876\n* fix(server): return plain text for response_format=text by @David-Wu1119 in https://github.com/modelscope/FunASR/pull/3874\n* fix(runtime): merge spelled-out acronyms into one word in funasr_onnx and funasr_torch sentence_postprocess by @tanishqbaweja in https://github.com/modelscope/FunASR/pull/3869\n* fix(runtime): trim batch padding from CIF peaks before funasr_onnx and funasr_torch Paraformer timestamps by @tanishqbaweja in https://github.com/modelscope/FunASR/pull/3871\n* fix(postprocess): preserve VAD word boundaries with current-main CI by @LauraGPT in https://github.com/modelscope/FunASR/pull/3883\n* fix(postprocess): keep a space between Latin VAD segments in rich ","is_prerelease":false},"readme":true,"readme_zh":true,"news":[],"stale":false,"snapshot_at":"2026-11-16T09:23:14Z","history_start":"2026-09-06"}