Agent Vox Voice Bridge

Example Use Cases

• 1

  Native voice input/output integration via MCP protocol.

• 2

  WebSocket-based voice communication for frontend or custom apps.

• 3

  Simple CLI-based speech recognition and synthesis.

Description

Agent-vox is a lightweight local voice service that adds speech input and output capabilities to AI agents like Claude Code, Codex, and Gemini CLI. It does not include an LLM but acts as the 'ears and mouth' for agents, enabling them to understand spoken language and respond with synthesized speech. It supports multiple integration modes including MCP protocol, WebSocket, and CLI, and uses privacy-focused local ASR and TTS technologies.