FunASR MCP Server Integration
Open-source speech recognition toolkit for training, inference, streaming ASR, VAD, punctuation, speaker diarization pipelines, and OpenAI-compatible/MCP serving.
Technical Integration & Multi-Client Support
The FunASR MCP Integration translates REST paths, operational endpoints, and tool schemas into standardized Model Context Protocol JSON-RPC 2.0 messages. This allows AI assistants like Claude Desktop, Cursor IDE, VS Code (Cline/Roo Code), and Zed Editor to run tool queries and execute functions seamlessly.
Add stdio configuration block to claude_desktop_config.json.
Configure workspace root at .cursor/mcp.json or Settings -> MCP.
Insert server JSON payload into cline_mcp_settings.json.
Specification & Compatibility Table
| Property | Specification Detail |
|---|---|
| Target Integration | FunASR (un) |
| Directory Category | ai ml |
| Protocol Spec | JSON-RPC 2.0 (stdio) |
| Canonical Path | /mcp-server/un/ |
Frequently Asked Questions
How do I access the full JSON configuration for FunASR?
Click 'Open Full FunASR MCP Config' above to view the complete parameter schema, environment variable setup, and copy-pasteable JSON configs for Claude Desktop, Cursor, and VS Code.
Does FunASR require authentication secrets?
Authentication depends on upstream API requirements. Check the environment variable table on the detail page to view required API keys and header tokens.
Browse by Category
Explore MCP server integrations organized by platform and use case.