Whisper Transcribe

Overview

Whisper Transcribe is an MCP server that transcribes audio locally using faster-whisper or connects to the OpenAI Whisper API. It integrates directly into AI assistants, exposing transcription as a tool your model can invoke.

The server runs locally as a Python package installed via uvx whisper-transcribe-mcp and communicates over stdio. It requires an API key for OpenAI Whisper access if you choose that route. The latest release is v1.1.2 under the MIT license, with source code available at ZahiriNatZuke/whisper-transcribe-mcp on GitHub.

Whisper Transcribe appears in the AI & Agents category and works with Claude Desktop, Claude Code, Cursor, VS Code, Windsurf, Zed, and other MCP clients. If you maintain this project, claim your listing on mcp.site to verify ownership and keep information up to date.

Installation

// claude_desktop_config.json
{
  "mcpServers": {
    "whisper_transcribe": {
      "command": "uvx",
      "args": [
        "whisper-transcribe-mcp"
      ]
    }
  }
}