Open-source speech recognition toolkit for training, inference, streaming ASR, VAD, punctuation, speaker diarization pipelines, and OpenAI-compatible/MCP serving.
Python asraudiochineseemotion-recognitionfunasrmcp-servermultilingual-asropenai-compatible-apiparaformerpunctuationpytorchreal-time-asrspeaker-diarizationspeech-recognitionspeech-to-textstreaming-asrtranscriptionvllmvoice-activity-detectionwhisper-alternative
feat: support SenseVoice speaker diarization (#2706)
游雁10ac8cf39e7d
diff --git a/README.md b/README.mdindex 4330e05b7..91dc7eb1c 100644--- a/README.md+++ b/README.md@@ -34,7 +34,7 @@## What's new:-- 2026/05/19: Fun-ASR-Nano now supports speaker diarization. Use with `vad_model` + `spk_model` + `punc_model` to get per-sentence speaker labels. See [demo](examples/industrial_data_pretraining/fun_asr_nano/demo_spk.py).+- 2026/05/19: Fun-ASR-Nano and SenseVoice now support speaker diarization. Use with `vad_model` + `spk_model` + `punc_model` to get per-sentence speaker labels. See [Fun-ASR-Nano demo](examples/industrial_data_pretraining/fun_asr_nano/demo_spk.py), [SenseVoice demo](examples/industrial_data_pretraining/sense_voice/demo_spk.py).- 2025/12/15: [Fun-ASR-Nano-2512](https://github.com/FunAudioLLM/Fun-ASR) is an end-to-end speech recognition large model trained on tens of millions of hours real speech data. It supports low-latency real-time transcription and covers 31 languages.- 2024/10/29: Real-time Transcription Service 1.12 released, The 2pass-offline mode supports the SensevoiceSmal model;([docs](runtime/readme.md));- 2024/10/10:Added support for the Whisper-large-v3-turbo model, a multitasking model that can perform multilingual speech recognition, speech translation, and language identification. It can be downloaded from the [modelscope](examples/industrial_data_pretraining/whisper/demo.py), and [openai](examples/industrial_data_pretraining/whisper/demo_from_openai.py).@@ -231,6 +231,27 @@ Parameter Description:- `merge_vad`: Whether to merge short audio fragments segmented by the VAD model, with the merged length being `merge_length_s`, in seconds (s).- `ban_emo_unk`: Whether to ban the output of the `emo_unk` token.+#### SenseVoice with Speaker Diarization++```python+from funasr import AutoModel+from funasr.utils.postprocess_utils import rich_transcription_postprocess++model = AutoModel(+ model="iic/SenseVoiceSmall",+ vad_model="fsmn-vad",+ vad_kwargs={"max_single_segment_time": 30000},+ spk_model="cam++",+ punc_model="ct-punc",+ device="cuda:0",+)+res = model.generate(input="example.wav", cache={}, language="auto", use_itn=True,+ batch_size_s=60, merge_vad=True, merge_length_s=15)+# Result contains sentence_info with speaker labels+print(res[0]["sentence_info"])+# [{'text': '...', 'start': 750, 'end': 5550, 'timestamp': [...], 'spk': 0}]+```+#### Paraformer```pythondiff --git a/README_zh.md b/README_zh.mdindex cc4d662ad..162f9b5c1 100644--- a/README_zh.md+++ b/README_zh.md@@ -34,7 +34,7 @@ FunASR 希望在语音识别的学术研究和工业应用之间架起一座桥## 最新动态-- 2026/05/19: Fun-ASR-Nano 现已支持说话人分离。配合 `vad_model` + `spk_model` + `punc_model` 使用,可获得带说话人标签的逐句结果。参见 [demo](examples/industrial_data_pretraining/fun_asr_nano/demo_spk.py)。+- 2026/05/19: Fun-ASR-Nano 和 SenseVoice 现已支持说话人分离。配合 `vad_model` + `spk_model` + `punc_model` 使用,可获得带说话人标签的逐句结果。参见 [Fun-ASR-Nano demo](examples/industrial_data_pretraining/fun_asr_nano/demo_spk.py)、[SenseVoice demo](examples/industrial_data_pretraining/sense_voice/demo_spk.py)。- 2025/12/15: [Fun-ASR-Nano-2512](https://github.com/FunAudioLLM/Fun-ASR) 是一款基于数千万小时真实语音数据训练的端到端语音识别大模型。它支持低延迟实时转写,并涵盖 31 种语言识别功能。- 2024/10/29: 中文实时语音听写服务 1.12 发布,2pass-offline 模式支持 SensevoiceSmall 模型;详细信息参阅([部署文档](runtime/readme_cn.md))- 2024/10/10:新增加 Whisper-large-v3-turbo 模型支持,多语言语音识别/翻译/语种识别,支持从 [modelscope](examples/industrial_data_pretraining/whisper/demo.py)仓库下载,也支持从 [openai](examples/industrial_data_pretraining/whisper/demo_from_openai.py)仓库下载模型。diff --git a/examples/industrial_data_pretraining/sense_voice/demo_spk.py b/examples/industrial_data_pretraining/sense_voice/demo_spk.pynew file mode 100644index 000000000..e38a5c93f--- /dev/null+++ b/examples/industrial_data_pretraining/sense_voice/demo_spk.py@@ -0,0 +1,61 @@+#!/usr/bin/env python3+# -*- encoding: utf-8 -*-+# Copyright FunASR (https://github.com/alibaba-damo-academy/FunASR). All Rights Reserved.+# MIT License (https://opensource.org/licenses/MIT)++"""+SenseVoice with Speaker Diarization++This demo shows how to use SenseVoice with VAD + Speaker Model + Punctuation Model+to get per-sentence speaker labels.+"""++import torch+from funasr import AutoModel+from funasr.utils.postprocess_utils import rich_transcription_postprocess+++def main():+ model_dir = "iic/SenseVoiceSmall"+ device = (+ "cuda:0"+ if torch.cuda.is_available()+ else "mps" if torch.backends.mps.is_available() else "cpu"+ )++ model = AutoModel(+ model=model_dir,+ vad_model="fsmn-vad",+ vad_kwargs={"max_single_segment_time": 30000},+ spk_model="cam++",+ punc_model="ct-punc",+ device=device,+ )++ wav_path = f"{model.model_path}/example/zh.mp3"+ res = model.generate(+ input=wav_path,+ cache={},+ language="auto",+ use_itn=True,+ batch_size_s=60,+ merge_vad=True,+ merge_length_s=15,+ )++ # Print full text+ text = rich_transcription_postprocess(res[0]["text"])+ print("Text:", text)+ print()++ # Print per-sentence results with speaker labels+ print("Speaker Diarization Results:")+ for sent in res[0]["sentence_info"]:+ sent_text = rich_transcription_postprocess(sent["text"])+ print(+ f" Speaker {sent['spk']}: [{sent['start']}ms - {sent['end']}ms] {sent_text}"+ )+++if __name__ == "__main__":+ main()diff --git a/funasr/auto/auto_model.py b/funasr/auto/auto_model.pyindex ddf19cc8c..dc7550c30 100644--- a/funasr/auto/auto_model.py+++ b/funasr/auto/auto_model.py@@ -442,6 +442,8 @@ class AutoModel:def inference_with_vad(self, input, input_len=None, **cfg):self._reset_runtime_configs()+ if self.spk_model is not None and "output_timestamp" not in cfg:+ cfg["output_timestamp"] = Truekwargs = self.kwargs# step.1: compute the vad modeldeep_update(self.vad_kwargs, cfg)