The say-mcp-server is an MCP server that adds text-to-speech functionality by leveraging the built-in say command on macOS.
To begin using it, install the server via npm and add it to your MCP client configuration. You can then call the speak tool to convert any text into speech.
Its key features include using native macOS speech synthesis, supporting extensive customization of voice, rate, and volume, and allowing speech to be processed in the background while you perform other tasks.
Practical uses include reading documents or web content aloud, providing audio feedback in applications, and building interactive voice-responsive tools.
You need macOS and Node.js version 14.0.0 or higher to run the server. The voice, speaking speed, and volume are all customizable, and background speech execution is fully supported.