elevenlabs-mcp

Why we included this project

For anyone running an MCP client who wants the assistant to actually speak, this is the official route into ElevenLabs' speech APIs. Claude Desktop, Cursor, Windsurf, and similar clients can call text-to-speech, voice cloning, speech-to-speech conversion, transcription, and other audio tools without you wiring up the API by hand. Setup is quick: add an API key, register the server in your client's config, and you are testing, with a free tier of 10k credits a month to start. It also covers the less glamorous parts of a real pipeline, like configurable file output, base64 resource mode for containerized or serverless runs, and a data-residency region. Since ElevenLabs maintains it, the tool surface stays aligned with the upstream API rather than drifting.