ElevenLabs MCP Server
Generate professional, high-fidelity audio using ElevenLabs directly from Claude. Perfect for creating voiceovers, audiobooks, podcasts, and more.
Features
- Professional Quality - Industry-leading text-to-speech
- Multiple Voices - Access to a wide variety of AI voices
- Voice Cloning - Use your own custom voices
- Multi-language - Support for 29+ languages
Requirements
- Node.js 18+
- ElevenLabs API key (free tier available)
Installation
Step 1: Get an API Key
- Sign up at ElevenLabs
- Go to your Profile Settings
- Copy your API key
Step 2: Configure Claude Desktop
Add the following to your claude_desktop_config.json:
json
{
"mcpServers": {
"elevenlabs": {
"command": "npx",
"args": ["-y", "@mamertofabian/elevenlabs-mcp-server"],
"env": {
"ELEVENLABS_API_KEY": "your-api-key-here"
}
}
}
}WARNING
Replace your-api-key-here with your actual ElevenLabs API key.
Step 3: Restart Claude Desktop
Close and reopen Claude Desktop to load the new MCP server.
Usage Examples
Once configured, you can ask Claude things like:
- "Generate speech saying 'Hello, welcome to my podcast'"
- "Create an audiobook narration for this text"
- "Read this article aloud with a British accent"
- "Generate audio for my YouTube video script"
Available Tools
The MCP server exposes these tools to Claude:
| Tool | Description |
|---|---|
text_to_speech | Convert text to spoken audio |
list_voices | List available voices |
get_voice | Get details about a specific voice |
Voice Selection
ElevenLabs offers many pre-made voices. Some popular options:
- Rachel - Calm, professional female voice
- Adam - Clear, neutral male voice
- Bella - Warm, friendly female voice
- Antoni - Authoritative male voice
API Limits
ElevenLabs free tier includes:
- 10,000 characters/month
- Access to pre-made voices
Paid plans offer more characters and voice cloning features.
Troubleshooting
Invalid API key
- Double-check your API key is correct
- Ensure there are no extra spaces
- Regenerate your key if needed
Audio not playing
- Check if the audio file was created successfully
- Verify your system's audio output is working
- Try a shorter text sample first