SenseVoice
STT.ai does not run this model. This page is reference information about it.
Transcribe with the models we do run →
5.5%
WER
50
Languages
50.0x
Speed
MIT
License
About SenseVoice
SenseVoice is a speech foundation model from FunAudioLLM that goes beyond transcription. It supports 50+ languages and includes capabilities for emotion recognition, audio event detection, and inverse text normalization in a single model.
Model Info
- ProviderFunAudioLLM
- Architecture-
- LicenseMIT
- UpdatedMar 2026
Frequently Asked Questions
SenseVoice is a speech-to-text model by FunAudioLLM. STT.ai does not currently run SenseVoice — this page is reference information about the model. For transcription on STT.ai, the model picker offers the Whisper family (Turbo, Large V3, Medium, Small).
On standard benchmarks, SenseVoice achieves around 5.5% Word Error Rate. Real-world accuracy depends on audio quality, accent, and language; for noisy or accented recordings, expect a few percentage points higher WER.
SenseVoice is not available on STT.ai. Its own licence terms govern running it yourself. STT.ai's free tier covers the Whisper models we do run — 600 minutes to start, then a monthly free top-up.
SenseVoice is released under MIT, a permissive open-source license. You can self-host SenseVoice on your own hardware or use our hosted version — both are commercially usable.
SenseVoice supports 50 languages. Auto-detection picks the right language for most audio; you can also specify it manually for a small accuracy lift.
SenseVoice processes audio at about 50.0x real-time on our GPUs. A 1-hour audio file finishes in under 1 minutes; longer files queue and notify by email when done.
SenseVoice has 234M parameters. Larger models tend to be more accurate but slower; STT.ai hosts SenseVoice on GPU so the parameter count doesn't affect your client-side performance.
SenseVoice accepts every format STT.ai supports — MP3, WAV, M4A, FLAC, OGG, MP4, MKV, MOV, WebM, AVI, and others. Output as TXT, SRT, VTT, DOCX, JSON, or PDF.
Yes. Speaker diarization runs alongside SenseVoice for every transcription — each speaker is labeled and you can rename them in the editor afterwards.
Yes. SenseVoice runs in our managed environment — audio is processed and deleted by default and never used for training without explicit opt-in. Pro plans add client-side encryption for transcripts at rest.
Use the compare-stt tool to run SenseVoice against any other supported model on the same audio — you'll see WER, segment count, speaker labels, and confidence scores side-by-side. The SenseVoice vs Whisper Large V3 comparison is the most commonly run.
Yes. Specify "sensevoice" as the model parameter on the /v1/transcribe endpoint. Python and Node.js SDKs include SenseVoice examples. Free API tier includes 100 minutes/month.
Yes. Because SenseVoice is MIT-licensed, you can self-host it. STT.ai's open-source page lists the project repo and weights. Most production teams use our hosted version to skip GPU procurement, model swaps, and ops.