STT-06Open source
Whisper
OpenAI's open-source speech recognition model that runs on your computer in many languages.
Open Whisper Opens the official site in a new tab. This page stays open so you can come back.
What is Whisper?
Whisper is OpenAI's open-source speech recognition model. You can run it on your computer to transcribe and translate audio in many languages, completely privately.
Used for
OpenAI's open-source speech recognition model that runs on your computer in many languages.
Best for
Private offline transcription
Price
Whisper is open source and free to download and run on your own computer.
How to start for free
- Install Python, then install Whisper from GitHub.
- Run it on an audio file from the command line.
- Or use a free desktop app built on Whisper.
Strengths
- Free and private
- Many languages
Limits to know
- Requires some technical setup
- Slow without a good graphics card
Learn more
Free plans and limits change often. Check the current details on the official Whisper website before relying on it for important work. Last reviewed October 2026.