Starstruk Logo MarkStarstruk

v1.0.0

badges/shieldsv1.0.0May 25, 2026by LauraGPT

AI Summary

The first official release of SenseVoice, a speech foundation model designed for multilingual speech understanding.

Key Highlights

  • Multilingual ASR supporting 50+ languages
  • Speech Emotion Recognition capabilities
  • Ultra-fast inference speeds (70ms)
  • Speaker Diarization support
  • Audio Event Detection features

New Features

  • Multilingual ASR
  • Speech Emotion Recognition
  • Audio Event Detection
  • Ultra-fast inference
  • Speaker Diarization

Full Release Notes

## SenseVoice v1.0.0

The first official release of SenseVoice, a speech foundation model for multilingual speech understanding.

### Highlights

- **Multilingual ASR** — 50+ languages, superior to Whisper on Chinese and Cantonese
- **Speech Emotion Recognition** — Happy, Sad, Angry, Neutral detection
- **Audio Event Detection** — Background music, applause, laughter, crying, coughing
- **Ultra-fast inference** — Non-autoregressive, 70ms for 10 seconds of audio (15x faster than Whisper)
- **Speaker Diarization** — Works with FunASR's VAD + SPK pipeline for who-said-what

### Quick Start

```python
from funasr import AutoModel

model = AutoModel(model="iic/SenseVoiceSmall", device="cuda")
result = model.generate(input="audio.wav")
print(result[0]["text"])
```

### Models

| Model | Languages | Parameters | Download |
|-------|-----------|------------|----------|
| SenseVoice-Small | 5 (zh/en/ja/ko/yue) | 234M | [ModelScope](https://modelscope.cn/models/iic/SenseVoiceSmall) · [HuggingFace](https://huggingface.co/FunAudioLLM/SenseVoiceSmall) |

### Links

- Paper: [FunAudioLLM](https://arxiv.org/pdf/2407.04051v3)
- Demo: [ModelScope](https://modelscope.cn/studios/iic/SenseVoice) · [HuggingFace](https://huggingface.co/spaces/FunAudioLLM/SenseVoice)
- Toolkit: [github.com/modelscope/FunASR](https://github.com/modelscope/FunASR)
- Website: [www.funasr.com](https://www.funasr.com)