# Voicemode

> Voicemode is an open-source MCP server for developer tools. Voice conversations with Claude Code and other MCP agents using local or cloud speech. It runs locally on your machine and is started by your AI client.

- **Kind:** MCP server
- **Pricing:** Open source
- **Platforms:** macOS, Linux, Windows, CLI, self-hosted
- **Repository:** https://github.com/mbailey/voicemode
- **MCP Registry name:** dev.voicemode/voicemode
- **Categories:** [developer tools](https://ai-powered.com/categories/developer-tools), [productivity](https://ai-powered.com/categories/productivity)
- **Listing updated:** 2026-11-18

## About

The README frames voice as a way to stay available when typing is not practical, such as walking, cooking or giving your eyes a rest. The agent speaks its replies and listens through your microphone, and smart silence detection stops recording when you stop talking.

Speech can run entirely on your machine with Whisper.cpp for speech-to-text and Kokoro for text-to-speech, which expose the same API as OpenAI so the server switches between local and cloud. Local services are preferred by default, and an OpenAI key serves as a fallback. Settings include audio format, Whisper model size, a text-only mode and optional saving of audio for debugging.

It installs as a Claude Code plugin or through a Python installer package, and runs on Linux, macOS, Windows (native or WSL) and NixOS with Python 3.10 to 3.14. The permissions section names two tools, converse and service, that can be pre-approved to avoid prompts.

## Features

- **Spoken conversations:** Speaks agent replies and listens through the microphone.
- **Local speech services:** Runs Whisper.cpp and Kokoro for offline use.
- **Cloud fallback:** Uses an OpenAI key when local services are unavailable.
- **Silence detection:** Stops recording when you stop speaking.
- **Configurable audio:** Sets format, Whisper model size and text-only mode.
- **Claude Code plugin:** Installs through the plugin marketplace.

## Good for

- Talk through a coding task while walking or cooking
- Rest your eyes while an agent works
- Dictate instructions without typing
- Keep voice processing fully local for privacy

## Install

### Claude Code (Terminal)

```sh
claude mcp add voicemode --env OPENAI_API_KEY='<OPENAI_API_KEY>' -- uvx voice-mode
```

### Claude Desktop (claude_desktop_config.json)

```json
{
  "mcpServers": {
    "voicemode": {
      "command": "uvx",
      "args": [
        "voice-mode"
      ],
      "env": {
        "OPENAI_API_KEY": "<OPENAI_API_KEY>"
      }
    }
  }
}
```

### Cursor (~/.cursor/mcp.json)

```json
{
  "mcpServers": {
    "voicemode": {
      "command": "uvx",
      "args": [
        "voice-mode"
      ],
      "env": {
        "OPENAI_API_KEY": "<OPENAI_API_KEY>"
      }
    }
  }
}
```

### VS Code (.vscode/mcp.json)

```json
{
  "servers": {
    "voicemode": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "voice-mode"
      ],
      "env": {
        "OPENAI_API_KEY": "<OPENAI_API_KEY>"
      }
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

```json
{
  "mcpServers": {
    "voicemode": {
      "command": "uvx",
      "args": [
        "voice-mode"
      ],
      "env": {
        "OPENAI_API_KEY": "<OPENAI_API_KEY>"
      }
    }
  }
}
```

### Other clients (your client's MCP settings (most use this shape))

```json
{
  "mcpServers": {
    "voicemode": {
      "command": "uvx",
      "args": [
        "voice-mode"
      ],
      "env": {
        "OPENAI_API_KEY": "<OPENAI_API_KEY>"
      }
    }
  }
}
```

## Tools

- `converse`: Starts a spoken conversation turn.
- `service`: Manages local voice services.

## FAQ

### Does it work offline?

Yes, with the optional local Whisper.cpp and Kokoro services.

### Do I need an OpenAI key?

Only if you want cloud speech; it is optional when local services are installed.

### Which platforms are supported?

Linux, macOS, Windows (native or WSL) and NixOS.

### What do I need?

A computer with a microphone and speakers.

### Is Voicemode free?

Yes. Voicemode is open source, so you can use and self-host it for free.

### How do I add Voicemode to Claude Code?

Run this command in your terminal: claude mcp add voicemode --env OPENAI_API_KEY='<OPENAI_API_KEY>' -- uvx voice-mode

### What are the best alternatives to Voicemode?

The closest alternatives to Voicemode are REA, Context7, Chrome DevTools MCP, Codebase Memory and GitHub.

## Where this comes from

- Description: written from the project's README.
- Install steps: from the official MCP Registry entry.

## Alternatives

- [REA](https://ai-powered.com/mcp/rea): REA is an open-source MCP server for developer tools.
- [Context7](https://ai-powered.com/mcp/context7): Context7 is a freemium MCP server for developer tools.
- [Chrome DevTools MCP](https://ai-powered.com/mcp/chrome-devtools-mcp-2): Chrome DevTools MCP is an open-source MCP server for browser automation.
- [Codebase Memory](https://ai-powered.com/mcp/codebase-memory): Codebase Memory is an open-source MCP server for knowledge and memory.
- [GitHub](https://ai-powered.com/mcp/github-2): GitHub is an open-source MCP server for developer tools.
- [Serena MCP: the IDE for your agent](https://ai-powered.com/mcp/serena-mcp-the-ide-for-your-agent): Serena MCP: the IDE for your agent is an open-source MCP server for developer tools.
- [awesome-mcp-servers](https://ai-powered.com/mcp/awesome-mcp-servers-2): awesome-mcp-servers is an open-source MCP server for developer tools.
- [XERJ](https://ai-powered.com/mcp/xerj): XERJ is an open-source MCP server for knowledge and memory.

---
Source: https://ai-powered.com/mcp/voicemode
