FastMCP FastMCP
Sign In
V

Voice Hooks

Real-time voice interaction system for Claude Code that enables natural speech-based conversations d...

5 views
0 installs
Updated Dec 20, 2025
Not audited
Communication Developer Tools
Real-time voice interaction system for Claude Code that enables natural speech-based conversations during development workflows with browser-based speech recognition, text-to-speech responses, and conversation state management.

  1. Click the "Install Now" button to open the FastMCP connection / environment variables interface.

  2. In the FastMCP environment variables form, add the first key:

    • Key: MCP_VOICE_HOOKS_PORT
    • Value: (choose a port number; default is 5111).
      • Recommendation: use 5111 if available. If port 5111 is in use, pick another unused port (for example, 8080).
      • Note: this value must match the port the MCP server and the Claude Code hooks will use.

  3. Add the second key:

    • Key: MCP_VOICE_HOOKS_AUTO_OPEN_BROWSER
    • Value: "false" to disable the automatic browser open; omit or set to "true" to keep the automatic open behavior.
    • Note: values are strings (e.g., "false").

  4. Save / Apply the environment variable changes in the FastMCP interface.

  5. Restart the MCP server / Claude Code so the new ENV values are picked up:

    • If running via MCP-managed mode, restart Claude Code (e.g., stop and start the Claude/MCP process).
    • Confirm the MCP server logs show it is listening on the port you set.

  6. (Optional) If you prefer to set these in your project file instead of the FastMCP UI, add them to .claude/settings.local.json in your project (example):

    {
  "env": {
    "MCP_VOICE_HOOKS_PORT": "5111",
    "MCP_VOICE_HOOKS_AUTO_OPEN_BROWSER": "false"
  }
}

    Save and restart Claude Code thereafter.

  7. No API keys are required for Voice Mode (voice recognition and text-to-speech are handled by the browser). Once the ENV values are set and the server restarted, open the browser interface (http://localhost:) and use the voice-hooks UI.

Quick Start

View on GitHub

More for Communication

View All →

More for Developer Tools

View All →

Similar MCP Servers

ElevenLabs

ElevenLabs

Official

Unleash powerful Text-to-Speech and audio processing with the official ElevenLabs MCP server. It enables MCP clients like Claude Desktop, Cursor, and OpenAI Agents to generate speech, clone voices, transcribe audio, and create unique sounds effortlessly. Customize voices, convert recordings, and build immersive audio scenes with easy-to-use APIs designed for creative and practical applications. This server integrates seamlessly, expanding your AI toolkit to bring rich, dynamic audio experiences to life across various platforms and projects.

Entertainment and Media AI and Machine Learning
396
6
V

Voice Interface

Provides browser-based voice input/output capabilities for conversations, featuring real-time speech-to-text recognition, text-to-speech synthesis, and voice message queuing through a web interface for hands-free interactions and accessibility applications.

Communication AI and Machine Learning
135
0
1-Click Ready

Report Issue

Thank you! Your issue report has been submitted successfully.