Whisper Mcp

jwulff/whisper-mcp

Adds local audio transcription through whisper.cpp with three straightforward tools: transcribe_audio for converting speech to text, list_whisper_models to see what's available, and download_whisper_model to pull down the model you need. Choose from tiny to large models depending on your speed versus quality tradeoff. Works with wav, mp3, m4a and other formats, returning plain text, timestamped output, or JSON. Pairs naturally with apple-voice-memo-mcp if you want to transcribe voice memos straight from your Mac. Everything runs locally, no API calls. Requires whisper.cpp and ffmpeg via Homebrew, works on macOS with Apple Silicon.

GitHub stars: 2
tools: 3
transport: STDIO
last updated: Jan 12, 2026

View on GitHub

Tools

3 tools

Public tool metadata for what this MCP can expose to an agent.

transcribe_audioTranscribe an audio file using Whisper.

Transcribe an audio file using Whisper.

No parameter schema in public metadata yet.

list_whisper_modelsList available Whisper models and their download status.

List available Whisper models and their download status.

No parameter schema in public metadata yet.

download_whisper_modelDownload a Whisper model for local use.

Download a Whisper model for local use.

No parameter schema in public metadata yet.

Readme

Whisper MCP Server

A lightweight MCP (Model Context Protocol) server for local audio transcription using whisper.cpp. There are several Whisper MCP implementations out there. This one is minimal and pairs with apple-voice-memo-mcp for a complete voice memo workflow.