MCPHIVE

ElevenLabs MCP

Local · stdio

ElevenLabs MCP is ElevenLabs' official server for generative audio: text-to-speech, voice cloning and voice design, transcription and audio track isolation. It runs locally over stdio via uvx or pip, authenticated with an ELEVENLABS_API_KEY. Output lands as a file on disk, an MCP resource, or both. The free tier includes 10,000 credits a month.

Why this server

  • uvx elevenlabs-mcp starts with one command — synthesis, cloning and transcription right away
  • Voice design creates a new voice from a text description, not just a clone of a sample
  • Audio isolation strips background noise and music with a single tool call
  • The free tier includes 10,000 credits a month — enough for testing and small tasks

Usage examples

Prompt: Read this text aloud in Rachel's voice and save it as a file

Result: The agent called text-to-speech with the chosen voice and saved the mp3 to disk via the configured output mode.

Prompt: Transcribe this call recording with speaker separation

Result: The transcription tool recognized the speech and returned text with speaker labels, no third-party service needed.

Local server

This MCP runs locally on the user's machine (stdio), so availability isn't tracked — rely on reviews and the rating instead.

Reviews

No reviews yet.