Loading...
437+ professionals trust Whisper
You hold a key and talk, and the text goes into whatever program you have open. It runs on your computer and nothing gets uploaded.
macOS (Apple Silicon) · Windows 10/11 · Linux (Ubuntu 22.04+)
You answer an email while you walk around the room. You finish the chart note before the next patient comes in. And an hour of recorded audio takes about a minute.
For teams
One license key for every device. Invoice with company name and VAT ID, bank transfer on request. Each extra device costs $29 a year.
And it cleans up what you said. The ums go, the repeats go, and the punctuation is already there.
"Um", "uh", "like", "you know"—gone. Your speech becomes clean, professional text.
So, um, I was thinking we could, like, move it to tomorrow?
I was thinking we could move it to tomorrow?
Said something twice? Whisper catches it and keeps only what matters.
I think we should, should probably send the report tomorrow... yeah tomorrow.
I think we should probably send the report tomorrow.
Punctuation, capitalization, paragraphs—all handled. Ready to send.
hey can you send me the file also i need it by friday thanks
Hey, can you send me the file? Also, I need it by Friday. Thanks!
All the cleanup runs locally with a built-in AI model. No account, no API key, no internet, and your voice never leaves your computer.
Learn how it workssend me the file now
Could you please send me the file when you have a moment? Thanks!
Speak in any language, switch mid-sentence — and translate your dictation into another language with your own AI key (optional).
Bonjour, ça va? Hola, ¿cómo estás? 你好吗？
Bonjour, ça va? Hola, ¿cómo estás? 你好吗？
No, nothing you say gets uploaded. It all runs on your computer.
Yes. Once the speech model is downloaded you don't need internet, so on a plane it works the same.
Mac with Apple Silicon, Windows 10 and 11, Linux (Ubuntu 22.04 or newer).
Write to hello@get-whisper.com and tell us what happens. We reply within one business day and help you get it set up.
$39 per year for 1 device. Each extra device adds $29 per year.
How many devices do you need?
device
Buying for a team? First device $39, each extra device $29 per year. See team pricing
Renews every year at $39 (or the same amount in your currency). We email you 7 days before. Cancel any time in your account or by replying to any email.
System requirements
macOS
Apple Silicon (M1-M4)
Windows 10 & 11
Windows 10/11
Linux Ubuntu 22.04+
Ubuntu 22.04+
Yes. Speech is turned into text on your computer, and the audio is never uploaded. The app only contacts our server to download speech models, check your license key and look for updates. Optional AI text cleanup through a cloud service only works with your own API key, and it is off by default.
It's good. Our own engine runs open-source speech models on your computer, and they handle accents and normal background noise well.
Write to hello@get-whisper.com and tell us what happens. We reply within one business day and help you get it set up.
Mac with Apple Silicon (M1/M2/M3/M4), Windows 10/11 PC, or Linux (Ubuntu 22.04+). A dedicated GPU helps but isn't required—we have CPU-optimized models too.
Whisper supports 99+ languages including English, Spanish, French, German, Chinese, Japanese, Korean, Arabic, Hindi, and many more. It handles accents and dialects remarkably well.
Absolutely! It's perfect for transcribing meetings, interviews, lectures, or any spoken content. Many users record audio and transcribe it afterwards for notes.
Yes! The Whisper AI model is trained on diverse speech patterns and handles various accents, dialects, and speaking styles with high accuracy.
Yes! You can set any key combination you want. Most users prefer a modifier key like Option or Ctrl plus another key for quick access.
Hiroshi T. from Tokyo
just purchased Whisper
2 minutes ago