Launching today

WhisperJot
Voice dictation that never leaves your machine
0 followers
Voice dictation that never leaves your machine
0 followers
Hold a hotkey, say the thing, and clean punctuated text lands wherever your cursor already is — email, editor, chat. No window to switch to, no paste step, no cleanup pass. The speech model runs on your own machine by default: zero outbound requests during a local dictation, and it keeps working with the Wi-Fi off. An optional cloud engine is one switch away. macOS, Windows and Linux (beta) on one account. Unlimited dictation, no meters. From $5/month, or $199 once for lifetime.








Payment Required
Launch Team

Framer AI AgentsDesign and publish professional sites with AI
Promoted
Hi Product Hunt 👋
I look after WhisperJot's communications and I'm the one who hunted it, so I'll be your point of contact here all day. Worth saying up front: I'm not the engineer who built it. Where a question needs a precise technical answer, I'll get you the exact one rather than a close one.
WhisperJot exists because of a trade every dictation app makes without asking: to turn your voice into text, it uploads your voice.
For a grocery list, fine. For a patient note, a legal draft, a password read aloud, or an unreleased product name — not fine. And you can rotate a leaked password. You cannot rotate your voiceprint.
So WhisperJot runs the speech model on your own machine by default. Not as a Pro upsell, not as a setting buried three menus deep — as the default. Watch the network log during a local dictation and you'll see zero outbound requests. Turn the Wi-Fi off and it keeps working.
How it works: hold a hotkey, say the thing, let go. Clean punctuated text lands wherever your cursor already is — your email, your editor, your chat. Fillers removed, punctuation placed, your names and jargon spelled the way you expect. No window to switch to, no paste step, no tidy-up pass.
macOS, Windows and Linux on one account. The Linux build is worth calling out: no premium app in this category ships one. It's beta and we'll keep calling it beta, but it's real — a .deb, a tarball, and native arm64 builds.
Three things you'd find out anyway, so here they are first:
1. There's no free trial. The local engine downloads a speech model on first run, so a trial would mean shipping you a worse version or a download you didn't ask for. Instead there's a 30-second browser demo at whisperjot.com/demo — no signup — and a 14-day money-back guarantee.
2. That browser demo runs on Jot Cloud, the optional cloud engine, because a 500 MB model can't ship into a web page. The app you install does not work that way — it's local by default. Better said plainly than found later.
3. Apple Silicon only (no Intel Macs), Windows x64 only (no ARM64 yet), and there's no mobile app.
Pricing: $5/month local-only, $99/year for everything, or $199 once for lifetime — and the lifetime one includes a second lifetime license to give to someone else.
What we'd genuinely like feedback on: which app do you dictate into most, and where does it fall down? Per-app tone and vocabulary handling is what's being built next, and your answers beat our guesses.
I'm here all day.