FunASR
Industrial-grade speech recognition toolkit supporting over 50 languages, speaker separation, and emotion detection.
- Type
- MCP
- Transport
- http
- Open source
- Yes
- GitHub Stars
- ★ 18.5k
- Source
- mcp-github
- Repository
- github.com/modelscope/FunASR
Overview
FunASR is an industrial-grade speech recognition toolkit that supports over 50 languages and features speaker separation and emotion detection. It is 170 times faster than Whisper and provides an OpenAI-compatible API. With simple installation and invocation, it can be easily integrated into various AI applications. Ideal for developers and enterprises requiring high-performance speech processing capabilities.
Capabilities
- ▪Supports over 50 languages
- ▪Real-time speech recognition
- ▪Speaker separation
- ▪Emotion detection
- ▪Streaming processing
- ▪OpenAI-compatible API
Use cases
Setup
pip install torch torchaudio && pip install funasr
This information was compiled by AI from public sources and may contain inaccuracies — please refer to the source.
FAQ
How can I get started with FunASR quickly?
You can get started quickly via Colab or install locally and use the example code.
Which languages does FunASR support?
Supports over 50 languages, including Chinese, English, and more.
Do I need a GPU?
GPU is recommended for optimal performance, but CPU operation is also supported.
Related skills
gemini-cli
Gemini CLI is an open-source AI agent that brings Gemini's powerful capabilities directly into the terminal.
private-gpt
Provides a complete API layer for local models, supporting RAG, skills, tools, and more.
planning-with-files
File-based persistent planning skill, ideal for AI coding agents and long-running tasks.
scientific-agent-skills
Transform any AI agent into a scientific assistant with 147 ready-to-use research skills.
susi_alexa_skill
A skill that enables question-and-answer interactions between Alexa and Susi AI.
claude-context
Provides code search capabilities for Claude Code, turning the entire codebase into context.