home / mcp / elevenlabs mcp server

ElevenLabs MCP Server

Provides an MCP server that converts text to speech via ElevenLabs API with multi-voice/script support and history storage.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "mcp-mirror-mamertofabian_elevenlabs-mcp-server": {
      "command": "uvx",
      "args": [
        "elevenlabs-mcp-server"
      ],
      "env": {
        "ELEVENLABS_STYLE": "0.1",
        "ELEVENLABS_API_KEY": "your-api-key",
        "ELEVENLABS_MODEL_ID": "eleven_flash_v2",
        "ELEVENLABS_VOICE_ID": "your-voice-id",
        "ELEVENLABS_STABILITY": "0.5",
        "ELEVENLABS_OUTPUT_DIR": "output",
        "ELEVENLABS_SIMILARITY_BOOST": "0.75"
      }
    }
  }
}

You can run the ElevenLabs MCP Server to generate audio from text using the ElevenLabs API, manage voices and multi-part scripts, and store a persistent history with SQLite. This server pairs with a web-based MCP Client for easy management and playback of voiceover tasks, making it straightforward to automate and reuse voice assets across projects.

How to use

You will run the MCP server locally or in your environment and connect an MCP Client to send text or scripts for speech synthesis. Use the client to create simple text-to-speech tasks or multi-part scripts with multiple voices and actors. You can review voiceover history, replay past jobs, and download generated audio files. All tasks are stored in a local database for easy history access.

How to install

Prerequisites: you need Node.js and a runtime environment that can run MCP servers, plus access to the ElevenLabs API.

Choose one of the two MCP server configurations shown below. Copy the JSON block that matches your setup into your MCP settings file used by your client.

{
  "mcpServers": {
    "elevenlabs": {
      "command": "uvx",
      "args": ["elevenlabs-mcp-server"],
      "env": {
        "ELEVENLABS_API_KEY": "your-api-key",
        "ELEVENLABS_VOICE_ID": "your-voice-id",
        "ELEVENLABS_MODEL_ID": "eleven_flash_v2",
        "ELEVENLABS_STABILITY": "0.5",
        "ELEVENLABS_SIMILARITY_BOOST": "0.75",
        "ELEVENLABS_STYLE": "0.1",
        "ELEVENLABS_OUTPUT_DIR": "output"
      }
    }
  }
}

Additional configuration notes

If you prefer a development setup, you can run the MCP server using a local directory path. Use the development command flow to run the server from your project workspace.

{
  "mcpServers": {
    "elevenlabs": {
      "command": "uv",
      "args": [
        "--directory",
        "path/to/elevenlabs-mcp-server",
        "run",
        "elevenlabs-mcp-server"
      ],
      "env": {
        "ELEVENLABS_API_KEY": "your-api-key",
        "ELEVENLABS_VOICE_ID": "your-voice-id",
        "ELEVENLABS_MODEL_ID": "eleven_flash_v2",
        "ELEVENLABS_STABILITY": "0.5",
        "ELEVENLABS_SIMILARITY_BOOST": "0.75",
        "ELEVENLABS_STYLE": "0.1",
        "ELEVENLABS_OUTPUT_DIR": "output"
      }
    }
  }
}

Available tools

generate_audio_simple

Generate audio from plain text using the default voice settings.

generate_audio_script

Generate audio from a structured script with multiple voices and actors.

delete_job

Delete a voiceover job by its ID.

get_audio_file

Retrieve the generated audio file by its ID.

list_voices

List all available voices supported by the ElevenLabs integration.

get_voiceover_history

Obtain the history of voiceover jobs; optionally narrow to a specific job by ID.