FunASR
OTR ACTIVE STEADY
modelscope/FunASR · homepage ↗
Industrial-grade speech recognition toolkit: 170x realtime, 50+ languages, speaker diarization, emotion detection, streaming, and OpenAI-compatible API.
- pytorch
- speech-recognition
- paraformer
- punctuation
- speaker-diarization
- voice-activity-detection
stars 19k
last activity yesterday
open issues 6
language Python
license MIT
latest release runtime-llamacpp-v0.1.4
momentum · per month since covered + 607/mo
(+6%/mo) · + 9k total since PR#7
metrics as of today
star history
Exact curve on star-history.com ↗- PR#7 10k★ 2025-04-30
- now 19k★ + 9k since first covered
curve is sampled from GitHub's star history; the dashed stretch is before we first covered it, the solid line since. figures at coverage are the numbers we printed then (approx.), current count is live.
covered in
-
A Comprehensive End-to-End Speech Recognition Toolkit
// comments
COMING SOONSign in with GitHub to weigh in on FunASR. We're wiring this up; check back soon.