Voice Interface

This MCP Voice Interface server provides browser-based voice input/output capabilities for AI Assistant conversations, featuring real-time speech-to-text recognition, text-to-speech synthesis, and voice message queuing through a web interface served on both HTTP and HTTPS ports. Built with Express, the MCP SDK, and Web Speech APIs, it offers tools for voice conversations through a 'converse' tool that speaks text and waits for voice responses, voice status monitoring, and message delivery management, while including an OpenCode plugin for automatic voice message forwarding to busy sessions. The implementation features self-signed certificate generation for HTTPS support, Server-Sent Events for real-time browser communication, configurable speech settings, and always-on microphone modes, making it valuable for hands-free AI interactions, accessibility applications, and voice-driven workflows where users need natural speech-based communication with AI assistants.

Source

Repository: https://github.com/shantur/jarvis-mcp