Whisper Mcp is a lightweight MCP (Model Context Protocol) server designed for local audio transcription workflows using OpenAI Whisper models via whisper.cpp. It provides easy, on-device audio-to-text transcription supporting multiple audio formats and a range of Whisper models (tiny, base, small, medium, large) with options for different output formats (plain text, timestamps, or JSON). It integrates natively with MCP-compatible assistants such as Claude Desktop, Claude Code, Cursor, and can pair with apple-voice-memo-mcp to enable a complete local voice memo transcription pipeline. It is ideal for developers, researchers, or privacy-focused users who require fast, local, and scriptable voice transcription without sending data to external servers.
Visit Whisper Mcp's official website for product details and getting started.