Windows and Android · 100% localTry Pro free — no card
Say it, and the text appears wherever you’re typing: chat, email, docs or code. It’s about 3× faster than typing, and it all runs on your device, with no internet.
On first launch Windows may show “Unknown publisher” — that’s expected. What to doOther options
90+ languages
Features
You speak about 120 words a minute and type about 40. Dictate into chat, email, docs and code.
Nothing to copy: recognized text is automatically inserted into the active window — Telegram, Word, your browser or IDE.
Will the contract be ready by tomorrow?
Yes, I’ll send the final version by noon — just two clauses left to align with legal.
dictated in 4 seconds
Three recognition models to choose from — Whisper, NVIDIA Parakeet and GigaAM. Each downloads in one click; nothing to configure.
“please move the meeting to three p m”
Please move the meeting to 3:00 PM.
Removes “um” and filler words, fixes punctuation, formats your text and translates it into dozens of languages. Runs on your PC — no internet needed.
“uh in the supply agreement we need to change the payment term from thirty days to forty five and also add a clause about a zero point one percent daily penalty”
Supply agreement: change the payment term from 30 to 45 days; add a clause on a 0.1% daily penalty.
With a graphics card, recognition takes a fraction of a second; without one, it runs on the CPU. NVIDIA, AMD or Intel — any Windows 10 or 11 PC will do.
Say “Kitty, undo” and your last action is undone. Shortcuts and clipboard commands are free; launching apps, opening links and multi-step actions are Pro.
Every transcription is stored locally: search, copy, see your stats. Nothing is sent to the cloud.
How it works
Hold your hotkey, speak, let go. The floating bar stays on top, always within reach.
Pauses are trimmed automatically, and the model transcribes your speech right on your computer.
The result is instantly inserted into the active input field — in any Windows app.
SweetWhisper turns your voice into text instantly. Fully offline, right on your PC.
Android
Dictate inside the app you are already writing in: Sweet Whisper lives in the keyboard.
Open a chat, tap the microphone on your keyboard, speak — the words land straight in the message. Nothing to copy, nowhere to switch. Speech is recognized on the phone itself: no internet, no account, and no recording ever leaves the device.
Dictating from the keyboard is free, with no limits. Pro adds the floating button, translation and the largest models — try them free for your first 200 floating-button dictations. One key unlocks Pro on Android and Windows.
Privacy
SweetWhisper is built local-first: audio is recorded, transcribed and discarded on your PC. No accounts, no telemetry, no “analytics to improve the service”. The app goes online for three main reasons: to download a model when you ask, to check for updates (you can turn this off), and to send error reports, but only if you turn them on. Anything else is something you start yourself, such as sending feedback, finding a server on your network, or using a cloud model; the privacy policy lists them all. Cloud engines (OpenAI, Yandex) are strictly opt-in, and their keys live in the protected Windows credential store.
Comparison
| Comparison | SweetWhisper | Wispr Flow | Open-source | Win+H |
|---|---|---|---|---|
| Works offline | ✓ | ✗ cloud | ✓ | ✗ for most languages |
| Audio stays on your computer | ✓ | ✗ | ✓ | ✗ |
| AI text cleanup | ✓ Pro, on-device | ✓ in the cloud | ✗ | ✗ |
| Voice commands | ✓ | text snippets | ✗ | ✗ |
Works offline
SweetWhisper✓
Wispr Flow✗ cloud
Open-source✓
Win+H✗ for most languages
Audio stays on your computer
SweetWhisper✓
Wispr Flow✗
Open-source✓
Win+H✗
AI text cleanup
SweetWhisper✓ Pro, on-device
Wispr Flow✓ in the cloud
Open-source✗
Win+H✗
Voice commands
SweetWhisper✓
Wispr Flowtext snippets
Open-source✗
Win+H✗
Testimonials
It just works: hold the key, speak — the text is already in the field. I use it daily instead of typing.
The design is a treat: a tidy floating bar, thoughtful themes. Windows apps are rarely this polished.
The feature set surprised me: voice commands, custom replacement dictionaries, history — didn’t expect that from local dictation.
What early testers told us — friends and colleagues who used SweetWhisper before the public launch. Paraphrased from memory.
Pricing
Dictation is free on Windows and Android — no limits, no time limit. Pro is a one-time purchase, and you can try it first: 14 days free on Windows, your first 200 floating-button dictations free on Android. One key unlocks Pro on both.
$0 forever
$19.99
One-time payment · yours forever · no subscription
Try it free: 14 days on Windows · 200 floating-button dictations on Android
What Pro unlocks:
Less than two months of a typical dictation subscription — and you pay once
| Free vs Pro in detail | Free | Pro |
|---|---|---|
| Unlimited local dictation | ✓ | ✓ |
| All recognition models and GPU speed | ✓ | ✓ |
| History, stats, basic voice commands | ✓ | ✓ |
| AI cleanup, punctuation and formatting | — | ✓ |
| Translate into any language (dictation & files; into English on Android) | — | ✓ |
| Task modes (writing, documents, code) | — | ✓ |
| Live text preview while dictating (needs a GPU) | — | ✓ |
| On-device AI text cleanup | — | ✓ |
| Cloud AI with your own API key (optional) | — | ✓ |
| Launch programs by voice, command chains | — | ✓ |
| Custom accent color | — | ✓ |
| Floating dictation button on Android (200 dictations free to try) | — | ✓ |
| Recordings transcribed with speaker separation | — | ✓ |
For teams
Put SweetWhisper on one powerful computer and the whole team dictates to it — from any device and from your own apps. Recognition runs on your node and never leaves your perimeter: neither the audio nor the text goes to any cloud.
Speech and text are processed on your server. Nothing leaves it — which matters for data under NDA and regulatory rules.
A single strong GPU serves recognition to dozens of devices and apps. No need to install a model on every machine.
A simple HTTP endpoint: your app sends audio, gets text back — with speaker separation on request. Access keys are issued and revoked in a couple of clicks.
The recognition server is part of the Corp plan. Tell us about your setup and we will come back with numbers for your team.
per node (one recognition server), per year
Licenses are annual. Your seat tier is set in the contract; the server itself doesn’t count or identify users.
Or simply write to hello@sweetwhisper.app
Thank you! We will reply to that address within one business day.
FAQ
Yes. Speech recognition runs locally on your computer — you only need internet once, to download a model. Cloud engines (OpenAI, Yandex SpeechKit) are optional; the default is fully offline.
No. Dictation is free, with no limits, forever. Pro is a one-time purchase that you keep for good: it adds AI cleanup, translation, task modes and more. Try it first: 14 days free on Windows, or your first 200 floating-button dictations on Android.
Yes. On Windows, every Pro feature is free for 14 days after you install. On Android there’s no time limit: you can use the Pro features — the floating button, translation and the largest models — for your first 200 floating-button dictations, and the app shows how many are left. No card, no sign-up. When the trial ends, Pro features lock until you add a key; regular dictation, including dictating from the keyboard on Android, stays free with no limits. One key unlocks Pro on both your computer and your phone.
Built-in dictation sends your voice to Microsoft’s cloud and needs internet for most languages. SweetWhisper works locally and offline, lets you choose the model, inserts text into any app and can clean up the result with AI (Pro).
Yes. SmartScreen shows that prompt for files few people have downloaded yet: reputation grows with each install, and every new version of an unsigned app starts from scratch. It warns about novelty, not malware. Click “More info” → “Run anyway”. To confirm the file is really ours, check its checksum — we publish them with every release, linked from the download section.
Google sends your speech to its servers; Sweet Whisper recognizes it on the phone itself, so it works on a plane or on the subway, collects nothing and shows no ads. It doesn’t replace your keyboard either: it steps in when you tap the microphone and hands you straight back.
Installer or portable build for Windows, APK for Android. No sign-up, no account.
Windows 10/11 · graphics card optional · v0.19.2
Windows: every Pro feature free for 14 days — no key, no card
This is Windows SmartScreen. It appears for files that few people have downloaded yet. Its reputation grows with each install, and every new version of an unsigned app starts from scratch. It does not mean the file is bad — it means it is new.
Don’t take our word for it: every release ships with checksums for both files. To check, run Get-FileHash on the downloaded file in PowerShell and compare the SHA-256 with the sum on the release page. Release checksums →
