What SenseVoiceSmall is
The model is described as a speech foundation model covering automatic speech recognition, spoken language identification, speech emotion recognition and audio event detection. The small variant uses a non-autoregressive end-to-end framework, with training on more than 400,000 hours of data and support for more than 50 languages.
What you can do with it
- Recognise speech and convert it into text
- Identify the spoken language in the audio
- Recognise emotion carried in the speech
- Detect audio events such as background music, applause and laughter
Model notes
- Trained on more than 400,000 hours of data
- Support for more than 50 languages
- The small variant handles 10 seconds of audio in 70 milliseconds
- The material states it is 15 times faster than Whisper-Large
Getting started
- Install the required libraries and dependencies such as funasr
- Import AutoModel and rich_transcription_postprocess
- Set the model path and device and initialise the model
- Call the generate method with an audio file path and a language parameter
Pros & cons
✓ What we like
- Fine-tuning scripts and strategies are provided
- A service deployment pipeline handles multiple concurrent requests
- Client languages include Python, C++, HTML, Java and C#
! What to watch out for
- The 70 millisecond figure is stated for 10 seconds of audio in the small variant
- Recognition is delivered through the documented model and deployment pipeline
- Audio event detection names background music, applause and laughter
FAQ
What abilities are listed?
Automatic speech recognition, spoken language identification, speech emotion recognition and audio event detection.
How fast is the small variant?
The material states 10 seconds of audio takes 70 milliseconds, 15 times faster than Whisper-Large.
Last reviewed: 2026-11-15
More AI audio tools tools
View all →-
Accent Guesser Accent recognition from a recording Free tools Creator tools Audio Speech to text Education tools -
AccurateScribe.ai Transcription for audio and video Free tools Creator tools Video to text Audio Speech to text -
Adobe Podcast Browser podcast recording and cleanup Free tools Creator tools Audio Speech to text Audio editing -
Agilotext French-first transcription and summaries Audio Speech to text Podcast Meeting transcription Meeting notes -
AI Dubbing Short video dubbing without signup Free tools Video translation Audio Voiceover Content localization -
AI Mastering Automatic online mastering Free tools Freemium tools Audio Podcast