mlx-whisper-mcp

mlx-whisper-mcp

3.4

If you are the rightful owner of mlx-whisper-mcp and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.

A simple Model Context Protocol (MCP) server that provides audio transcription capabilities using MLX Whisper on Apple Silicon Macs.

The MLX Whisper MCP Server is designed to facilitate audio transcription using the MLX Whisper model on Apple Silicon Macs. It supports transcribing audio files directly from disk, base64-encoded data, and YouTube videos. The server leverages the high-quality `mlx-community/whisper-large-v3-turbo` model to ensure accurate transcriptions. It is a self-contained script that manages dependencies automatically using `uv run`, providing a seamless setup experience. The server outputs rich console logs for debugging and saves transcription text files alongside the original audio files. It is optimized for use with Claude Desktop, offering both automated and manual configuration options.

Features

  • Transcribe audio files directly from disk
  • Transcribe audio from base64-encoded data
  • Download and transcribe YouTube videos
  • Uses the high-quality `mlx-community/whisper-large-v3-turbo` model
  • Self-contained script with automatic dependency management via `uv run`

Tools

  1. transcribe_file

    Transcribes an audio file from a path on disk.

  2. transcribe_audio

    Transcribes audio from base64-encoded data.

  3. download_youtube

    Downloads a YouTube video.

  4. transcribe_youtube

    Downloads and transcribes a YouTube video.