Free & open source · Windows 11 specific steps

Whisper AI Offline Install on Windows 11

Using winget and PowerShell to get OpenAI's Whisper running as a fully offline speech-to-text tool on Windows 11.

Get it from GitHub

Opens OpenAI's official Whisper repository

Maker

OpenAI

Cost

Free

Runs

Fully offline

Windows 11

Install using winget (built into Windows 11)

Open PowerShell as Administrator

Right-click the Start button → "Terminal (Admin)". Windows 11 ships with winget already installed.

Install Python and FFmpeg with winget

winget install Python.Python.3.11
winget install Gyan.FFmpeg

Close and reopen the terminal

This refreshes your PATH so python and ffmpeg are recognized commands.
Install Whisper with pip
```
pip install openai-whisper
```
Transcribe your first file
```
whisper "C:\Users\YourName\Desktop\audio.mp3" --model small
```
The transcript is saved as a .txt (plus .srt and .vtt) in the same folder.

Speeding it up with your GPU

NVIDIA GPU	Install CUDA-enabled PyTorch first — Whisper detects and uses it automatically
No dedicated GPU	Runs on CPU without any extra steps, just slower on larger models
Laptop with hybrid graphics	Set your NVIDIA GPU as preferred in Windows Graphics settings before running

Common Windows 11 issues

"whisper is not recognized" after install

Close and reopen the terminal, or restart your PC — Windows sometimes needs a fresh PATH read.
"Microsoft Visual C++ 14.0 required" error

Install the free "Build Tools for Visual Studio" from Microsoft — some Whisper dependencies need it to compile.
FFmpeg not found

If winget's install doesn't add it to PATH automatically, add FFmpeg's bin folder manually via System → Environment Variables.

Good to know: This is an independent guide, not affiliated with OpenAI or Microsoft. Every command above uses Windows 11's built-in package manager (winget) and pip — nothing is downloaded from a third-party mirror. All transcription happens locally; no audio ever leaves your PC.