Local Speech-to-Text MCP Server

Example Use Cases

• 1

  Fast, local speech-to-text transcription without relying on cloud services.

• 2

  Speaker diarization to identify and separate multiple speakers in audio.

• 3

  Support for various audio formats with automatic conversion and multiple output formats.

Description

This MCP server provides fully local speech-to-text transcription using whisper.cpp, ensuring complete privacy without cloud dependencies. It supports multiple audio formats with automatic conversion, speaker diarization, and outputs in various formats like txt, json, and srt. Optimized for Apple Silicon, it delivers fast transcription speeds with low memory usage, making it suitable for privacy-conscious and performance-demanding applications.