FunASR-Powered MCP Server

Example Use Cases

• 1

  Transcribe long audio files asynchronously with detailed timestamp information.

• 2

  Validate audio file integrity and properties before processing.

• 3

  Detect speech segments within audio files using voice activity detection.

Description

This MCP server leverages Alibaba's FunASR models to offer robust speech processing capabilities including audio file validation, asynchronous speech-to-text transcription with detailed timestamping, and voice activity detection. It supports dynamic loading and switching of ASR and VAD models, allowing flexible configuration per request or globally. The server is designed for extensibility and integrates with the FastMCP framework for easy interaction via MCP clients or HTTP.