Prompt With Jarvis
Work with meGet the free pack
All guides

NARRATEYou commented this keyword

ElevenLabs MCP

Let Claude generate speech, design and clone voices, transcribe audio and create sound effects through ElevenLabs.

Before you start:
  • ElevenLabs account (free tier available)
  • Claude Desktop or another MCP client
Setup
5 min
Level
Intermediate
Stars
1.5k
License
MIT
Jump to the install commandView on GitHub

What it does

This is the official ElevenLabs MCP server. An MCP server is a connector that gives Claude new tools. With it, Claude can turn text into speech, design and clone voices, transcribe recordings, and make sound effects.

Important: this local server is now deprecated and no longer actively maintained. ElevenLabs recommends its hosted MCP server instead, which needs nothing installed and signs you in with OAuth.

Install it (hosted, recommended)

  1. Create an ElevenLabs account if you do not have one.
  2. In Claude, add a new MCP connector with the server URL https://api.elevenlabs.io/v1/mcp.
  3. Sign in to ElevenLabs when asked. No API key is copied into your client.
  4. See ElevenLabs' hosted MCP documentation (linked below) for client-specific steps.

Or run the local server in Claude Desktop

  1. Get an API key from ElevenLabs. The free tier includes 10k credits per month.
  2. Install uv with: curl -LsSf https://astral.sh/uv/install.sh | sh
  3. In Claude Desktop go to Claude, Settings, Developer, Edit Config, and add the config below with your key.
  4. On Windows, turn on Developer Mode first: Help in the top-left menu, then Enable Developer Mode.

Local server config

Config
{
  "mcpServers": {
    "ElevenLabs": {
      "command": "uvx",
      "args": ["elevenlabs-mcp"],
      "env": {
        "ELEVENLABS_API_KEY": "<insert-your-api-key-here>"
      }
    }
  }
}

Use it

  1. Try: Generate three voice variations for a wise, ancient dragon character, then I will choose my favorite voice to add to my voice library.
  2. Try: Create a soundscape of a thunderstorm in a dense jungle with animals reacting to the weather.
  3. Try: Turn this speech into text, identify different speakers, then convert it back using unique voices for each person.

Good to know

These tools use ElevenLabs credits. With the local server, generated files save to your Desktop by default; set ELEVENLABS_MCP_BASE_PATH to change that folder. If you see the error spawn uvx ENOENT, run which uvx in your terminal and use that full path as the command.

Links

Want the next one first?

One new guide a week, straight to your inbox.

More like this