modelscope/FunASR
Some data may be outdated — see provenance labels.
Open-source speech recognition toolkit for training, inference, streaming ASR, VAD, punctuation, speaker diarization pipelines, and OpenAI-compatible/MCP serving. observed · 2026-10-02
Health v2 · maintenance only
93/100
- Activity 97
- Release rhythm 83
- Longevity 100
How is this computed?
round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10) — computed 2026-10-21. Adoption (stars, forks) is never an input.
- gap_med: 0.0
- age_days: 1427
- days_rel: 32
- days_push: 19
- n_releases_24m: 61
Adoption not part of the score
20568 stars · 2055 forks observed · 2026-10-02
What it is AI-extracted, prompt v1, taxonomy v1, 2026-10-19, confidence 0.99
FunASR is an open-source Python toolkit for speech recognition, offering offline, streaming, and edge deployment capabilities. It includes pipelines for VAD, punctuation, speaker diarization, and an OpenAI‑compatible API.
Use cases
- Transcribe audio files to text with high accuracy.
- Perform real‑time speech‑to‑text streaming via WebSocket.
- Detect voice activity and segment recordings automatically.
- Identify and diarize multiple speakers in a conversation.
- Add punctuation and formatting to raw ASR output.
- Deploy multilingual ASR models on edge devices using GGUF or llama.cpp.
When to choose
- You need a single Python library that covers ASR, VAD, punctuation, and speaker diarization.
- You want an OpenAI‑compatible API for speech recognition.
- Your project requires streaming transcription over WebSocket.
- You need multilingual support and the ability to run models on edge hardware.
When to avoid
- You require a non‑Python implementation or low‑level C/C++ library.
- Your use case demands ultra‑low latency inference that only specialized on‑device SDKs can provide.
- You need a solution that works without any GPU/accelerator support on very constrained devices.
Facets
library · maturity active
speech-recognition websocket streaming speech-processing python windows openai-compatible streaming edge-runtime multilingual vllm-acceleration gguf transformers voice-activity-detection speaker-diarization punctuation linux macos
2 sources
- readme: https://github.com/modelscope/FunASR · fetched 2026-10-02 · 9f4cb70947a2
- registry_pypi: https://pypi.org/pypi/funasr/json · fetched 2026-10-05 · d2e0f4e167b5
Member repositories
| Repository | Role | Health v2 |
|---|---|---|
| modelscope/FunASR | main | 93 |
For agents
markdown · JSON · MCP: product_card(name="modelscope/FunASR")
Data as of 2026-10-19T08:05:51.212177+00:00 · Report a problem