FunASR
ModelScope's speech recognition toolkit for offline, streaming and edge ASR, with VAD, punctuation, speaker diarization and OpenAI-compatible/MCP serving.
Updated
✨ Try AgentScout's free AI image generator →Good for: building AI agents, building voice agents, connecting tools with MCP
Use cases
About FunASR
FunASR is listed in our voice category and is aimed at builders. ModelScope's speech recognition toolkit for offline, streaming and edge ASR, with VAD, punctuation, speaker diarization and OpenAI-compatible/MCP serving. Pricing, as the maker states it: Open source. Maturity: Mature. We last checked the official link on 2026-10-18.
What is FunASR?
FunASR is an industrial speech-recognition toolkit from ModelScope for offline, streaming and edge deployment.
It covers speech recognition, voice activity detection, punctuation, speaker pipelines and emotion and audio-event models, and can serve models through an OpenAI-compatible API. Fun-ASR-Nano can also run through Hugging Face Transformers.
Sources: the official README and our directory listing, checked 2026-10-18. Official site: github.com/modelscope/FunASR.
Who is FunASR for?
Developers building voice agents or transcription features who want self-hosted speech recognition, including streaming and edge setups.
FunASR at a glance
- Open source: Yes (MIT License)
- GitHub: modelscope/FunASR · 20,784 stars (as of Oct 21, 2026)
Key features
- Speech recognition for offline, streaming and edge use
- Voice activity detection and punctuation
- Speaker, emotion and audio-event models
- OpenAI-compatible serving
- Hugging Face Transformers quickstart and Colab notebook
FunASR FAQ
What is FunASR?
FunASR is an industrial speech-recognition toolkit from ModelScope for offline, streaming and edge deployment.
Who is FunASR for?
Developers building voice agents or transcription features who want self-hosted speech recognition, including streaming and edge setups.
Is FunASR free?
FunASR's pricing, as the maker states it: Open source. Open-source tools are free to use, though you may still pay for model API usage or a hosted plan. Check the official site for current plans.
Is FunASR open source?
Yes. FunASR is released under the MIT License, according to its GitHub repository.
What are alternatives to FunASR?
Similar tools include Dograh, Speechmatics, TryVoice, Voiceloop, Hume AI and Patter.
FunASR ratings and reviews
Be the first to rate FunASR: tap 1 to 5 stars below. A written review is optional.
FunASR alternatives
Other voice tools in our directory. See all voice tools →
Dograh
Self-hostable voice agent platform with a visual call-flow builder, bring-your-own speech and LLM providers and MCP control, pitched as an open alternative to Vapi and Retell.
Speechmatics
Speech-to-text and voice AI APIs that handle accents, multiple speakers and multilingual speech for voice agents.
TryVoice
Hands-free voice runtime for coding agents with wake word, push-to-talk and streaming replies.
Voiceloop
Zero-dependency JS library for an interruptible in-browser voice loop (VAD, STT, LLM, TTS).
Hume AI
Voice AI company focused on emotionally intelligent speech, with voice-native evaluation and human judgment tooling for voice agents.
Patter
Open-source SDK that gives your AI agent a phone number, handling speech, real-time voice and telephony carriers in Python or TypeScript.
Popular searches with this tool
🏆 Ranked in: Best AI Voice Agent Platforms
🧭 Task guides: Voice and phone agents for building AI agents
Badge for makers
Built FunASR? Add this badge to your README or site. It links back to this page. About badges
