FunASR: Speech Recognition Demo

Industrial-grade ASR toolkit. Upload audio and get transcription instantly.

  • SenseVoice: Multi-task (ASR + emotion + events), 5 languages, ultra-fast
  • Paraformer: Non-autoregressive Chinese ASR with punctuation

GitHub | Docs | pip install funasr

Model

Install & Use Locally

pip install funasr
from funasr import AutoModel
model = AutoModel(model="funasr/paraformer-zh", hub="hf", vad_model="funasr/fsmn-vad", punc_model="funasr/ct-punc")
result = model.generate(input="audio.wav")