"Speech & Transcription" matching MCP servers:

MCP server providing 115 tools to control the Windy Word voice-to-text desktop app, enabling paste, hotkeys, transcription, recording, and many other operations.

Local Windows x64 MCP server for USB HID mouse control and optional voice dictation. Requires compatible uMouse hardware, firmware and x64 Node.js 22+. MCPB 0.1.1 packages the legacy 0.1.0 stdio runtime; shared-v2 is a separate source development path. The MCPB includes no cloud credentials, defaults speech recognition to disabled, and supports optional local Windows Speech.

Enhanced MCP server for sending native macOS notifications with terminal-notifier, interactive dialogs, text-to-speech, screenshots, and file selection.

macOS Notification MCP enables AI assistants to trigger native macOS sounds, visual notifications, and text-to-speech. Built for Claude and other AI models using the Model Context Protocol.

Enables direct, zero-token terminal streaming from LibreChat into host tmux sessions and provides FastMCP tools for AI agents to inspect and supervise those sessions.