Whisper Windows MCP Server

Windows-native MCP server for local audio transcription using whisper.cpp with Vulkan GPU acceleration

Local serverstdioGo

What is the Whisper Windows MCP MCP server?

Most AI and media services work still happens through a UI a human drives. Whisper Windows MCP MCP server moves it into the conversation instead. Windows-native MCP server for local audio transcription using whisper.cpp with Vulkan GPU acceleration.

The short version

A Windows-native MCP (Model Context Protocol) server that lets Claude Desktop transcribe audio and video files locally using whisper.cpp — with GPU acceleration, multilingual support, and batch processing. All transcription runs locally — no audio, video, or file paths ever leave your machine.

Getting it running

whisper-windows-mcp on npm is all you need. Most clients run it directly, so configuration is a few lines and a restart.

The tools it exposes

The server publishes 12 tools. What each one is for:

  • transcribe_audio — Transcribe a single file. Supports blocking (default) or background mode for long files
  • check_progress — Monitor a background transcription job started with transcribe_audio (background=true)
  • start_batch — Automated sequential batch transcription of all untranscribed files in a folder. Sorts by duration (shortest first), processes one at a time as
  • check_batch_progress — Monitor a running batch. Automatically advances to the next file when the current one finishes. Returns overall progress, current file with
  • generate_subtitles — Generate subtitle files. Supports automatic language detection and English translation output. Outputs SRT (widest compatibility) or WebVTT (web and
  • analyze_media — Analyze files before committing to transcription. Returns duration, size, codec, and estimated transcription time on CPU and GPU. For folders, shows
  • check_config — Verify whisper-cli.exe, the model file, and FFmpeg are all accessible. Run this first if anything is failing
  • list_models — List all Whisper model files installed in your models directory. Shows filename, size, whether it is currently active, quantization status, and
  • download_model — Download a Whisper model directly from Hugging Face into your models directory. Only downloads from trusted Hugging Face namespaces. After
  • switch_model — Switch the active Whisper model for the current session without restarting Claude Desktop. Change is session-scoped — does not persist after restart
  • check_system — Detect GPU hardware and verify Vulkan acceleration is available. Reports GPU name, VRAM, whether ggml-vulkan.dll is present, and recommends the best
  • whisper_server — Start, stop, or check the persistent model server (whisper.cpp's whisper-server). While running, the active model stays resident in VRAM and

What it needs from you

Configuration is passed through the environment: WHISPER_CLI_PATH, WHISPER_MODEL, WHISPER_SERVER_PATH, FFMPEG_PATH. Treat anything key-shaped as a real credential — scope it to the minimum the server needs, and rotate it if it ever lands in a shared config.

  1. Node.js 18 or laternodejs.org 2. whisper.cpp binaries with Vulkan GPU support — see Step 1 3. A Whisper model file — see Step 2 4. FFmpeg — required for video files and non-WAV/MP3 audio ---

Things to watch

  • It runs with your machine's permissions. That is convenient and also the reason to think about what you point it at before you approve a tool call.
  • With 12 tools registered it takes up a noticeable share of the context window; turn it off in projects that never touch Whisper Windows MCP.
  • Keep per-call confirmation enabled while you learn its behaviour; it is the cheapest safeguard you have.

How it compares

Plenty of AI and media services servers cover similar ground. The differences that matter in practice are scope of access and how much setup stands between you and a working tool call. Whisper Windows MCP's toolset — transcribe_audio, check_progress, start_batch and 9 more — is a fair guide to whether it matches your workflow. It is maintained by eviscerations; worth a glance at recent repository activity before you build anything load-bearing on it.

SyncDev reviews every entry in this directory against the project's own documentation before publishing, and revisits them as servers change.

Available tools

ToolWhat it does
transcribe_audioTranscribe a single file. Supports blocking (default) or background mode for long files.
check_progressMonitor a background transcription job started with transcribe_audio (background=true).
start_batchAutomated sequential batch transcription of all untranscribed files in a folder. Sorts by duration (shortest first), processes one at a time as background jobs, validates each output. Batch self-advances when each file f
check_batch_progressMonitor a running batch. Automatically advances to the next file when the current one finishes. Returns overall progress, current file with timestamp, and any failed files.
generate_subtitlesGenerate subtitle files. Supports automatic language detection and English translation output. Outputs SRT (widest compatibility) or WebVTT (web and HTML5 video).
analyze_mediaAnalyze files before committing to transcription. Returns duration, size, codec, and estimated transcription time on CPU and GPU. For folders, shows all files in a sortable table with transcription status.
check_configVerify whisper-cli.exe, the model file, and FFmpeg are all accessible. Run this first if anything is failing.
list_modelsList all Whisper model files installed in your models directory. Shows filename, size, whether it is currently active, quantization status, and recommended use case. No network calls — reads local filesystem only.
download_modelDownload a Whisper model directly from Hugging Face into your models directory. Only downloads from trusted Hugging Face namespaces. After downloading, use switch_model to activate it.
switch_modelSwitch the active Whisper model for the current session without restarting Claude Desktop. Change is session-scoped — does not persist after restart. To make permanent, update WHISPER_MODEL in your config.
check_systemDetect GPU hardware and verify Vulkan acceleration is available. Reports GPU name, VRAM, whether ggml-vulkan.dll is present, and recommends the best model size for your hardware.
whisper_serverStart, stop, or check the **persistent model server** (whisper.cpp's whisper-server). While running, the active model stays resident in VRAM and every transcribe_audio / transcribe_batch call is served over localhost wit

How to install the Whisper Windows MCP MCP server

{
  "mcpServers": {
    "whisper-windows": {
      "command": "npx",
      "args": ["-y", "whisper-windows-mcp"],
      "env": {
        "WHISPER_CLI_PATH": "your-value",
        "WHISPER_MODEL": "your-value",
        "WHISPER_SERVER_PATH": "your-value",
        "FFMPEG_PATH": "your-value"
      }
    }
  }
}

Add to claude_desktop_config.json, then restart Claude Desktop.

Configuration

  1. Node.js 18 or laternodejs.org 2. whisper.cpp binaries with Vulkan GPU support — see Step 1 3. A Whisper model file — see Step 2 4. FFmpeg — required for video files and non-WAV/MP3 audio ---
VariableDescriptionRequired
WHISPER_CLI_PATHFilesystem location the server is allowed to use.Optional
WHISPER_MODELConfiguration value read at startup.Optional
WHISPER_SERVER_PATHFilesystem location the server is allowed to use.Optional
FFMPEG_PATHFilesystem location the server is allowed to use.Optional

Example prompts to try

  • Use Whisper Windows MCP to transcribe audio.
  • Use Whisper Windows MCP to check progress.
  • Use Whisper Windows MCP to start batch.

Frequently asked questions

It connects Whisper Windows MCP to MCP-compatible AI assistants such as Claude and Cursor, exposing 12 tools (transcribe_audio, check_progress, start_batch, and more) that the assistant can call on your behalf. Instead of copying data back and forth by hand, the assistant works with Whisper Windows MCP directly.