ElevenLabs MCP Server

Overview

The official ElevenLabs MCP server is a local, open-source server that connects developers directly to the ElevenLabs AI audio platform. Whether you're using Claude, Cursor, or a custom script, you can build Conversational AI voice agents, perform outbound calls, transcribe speech, and generate audio - all with simple API calls.

Key Features

• Text-to-Speech: Convert text to natural-sounding speech
• Voice Cloning: Create custom voice models
• Audio Transcription: Transcribe speech to text
• Voice Agents: Build advanced voice agents that can perform outbound calls
• Complete Platform Access: Access to the entire ElevenLabs AI audio platform via simple text prompts

Setup

Getting started is straightforward:

1. Get your API key from ElevenLabs (free tier includes 10k credits per month)
2. Install uv (Python package manager)
3. Configure Claude Desktop's config file to include the server settings with your API key

Example Use Cases

• Create AI agents that speak like specific characters
• Generate voice variations
• Convert voice recordings
• Create soundscapes
• Transcribe speech and convert it back using unique voices for each person
• Build conversational voice agents
• Automated phone calls with AI voices

Integration

The official GitHub repository is available at github.com/elevenlabs/elevenlabs-mcp, and the server integrates seamlessly with:

• Claude Desktop
• Cursor
• Windsurf
• OpenAI Agents
• Any MCP-compatible client

Pricing

Free tier available with 10,000 credits per month. Paid plans for higher usage.