eviscerations/whisper-windows-mcp

What it does

whisper-windows-mcp is an MCP server that provides native Windows audio and video transcription capabilities powered by Whisper.cpp. It enables offline transcription without relying on cloud APIs, featuring Vulkan GPU acceleration for improved performance. The tool supports batch processing of multiple files, multilingual transcription, automatic model management, and background job handling for uninterrupted workflows.

How to set up

This is an MCP (Model Context Protocol) server designed for Windows environments. Installation involves: (1) Cloning or downloading the repository from GitHub, (2) Installing dependencies (Whisper.cpp and Vulkan drivers), (3) Configuring the MCP server settings for your AI client, (4) Running the server as a background service. Detailed setup instructions are available in the GitHub repository documentation.

Key features

• Local, offline transcription with no cloud API dependencies
• Vulkan GPU acceleration for faster processing
• Batch processing support for multiple audio/video files
• Multilingual transcription capabilities
• Automatic model downloading and management
• Background job handling for non-blocking operations
• Native Windows implementation without Python requirements
• MCP server integration for seamless AI tool integration