Microphone
Required to capture your voice. Audio is processed locally by default — it only goes to our servers if you select the optional Jot Cloud engine, and it's never stored there.
Local specs below · Jot Cloud needs none
System requirements
WhisperJot runs two ways, and only one of them cares about your hardware. The local engines — the default — transcribe on your own machine, fully private and offline, so they have the specs below. Jot Cloud, our opt-in cloud engine, has no minimum specs at all: it just needs an internet connection.
macOS
The primary, fully-supported platform.
| Component | Minimum | Recommended |
|---|---|---|
| Operating system | macOS 14 Sonoma | macOS 15 Sequoia or later |
| Processor | Apple Silicon (M1) Intel Macs not supported | M2 / M3 / M4 family |
| Memory (unified) | 8 GB | 16 GB+ required for local-LLM cleanup |
| Free disk space | ~3 GB app + one speech model | 10 GB+ room for multiple models |
| Microphone | Any built-in or external mic | |
| Internet | Only for the one-time model download (and the optional Jot Cloud engine, if you use it) — local dictation works fully offline | |
Windows
Native C#/.NET app for Windows 10 and 11.
| Component | Minimum | Recommended |
|---|---|---|
| Operating system | Windows 10 (64-bit, 22H2) | Windows 11 |
| Architecture | x64 ARM64 not yet supported | x64 |
| Processor | Modern quad-core CPU | Modern 8-core CPU speech models run on the CPU |
| Memory | 8 GB RAM | 16 GB+ RAM |
| Free disk space | ~3 GB | 10 GB+ |
| Runtime | Self-contained build — no separate .NET install required | |
| Internet | Only for the one-time model download (and the optional Jot Cloud engine, if you use it) — local dictation works fully offline | |
Linux
Shares the Windows port's .NET core — in active development.
| Component | Minimum | Recommended |
|---|---|---|
| Distribution | Any modern 64-bit distro .deb targets Debian/Ubuntu; the tar.gz runs anywhere | Ubuntu or Debian the most-tested path |
| Architecture | x64 | x64 or arm64 native arm64 builds ship too |
| Display server | Wayland or X11 hotkey capture via the XDG portal (GNOME 48+, KDE Plasma, Hyprland), evdev, or a CLI trigger socket | |
| Runtime libraries | libgomp1 and libasound2 (or libasound2t64) — pulled in automatically by the .deb |
|
| Hardware | Same speech models as macOS and Windows — see the model-size and local-LLM RAM notes below | |
| Internet | Only for the one-time model download (and the optional Jot Cloud engine, if you use it) — local dictation works fully offline | |
Below the specs? Jot Cloud runs on anything.
Our cloud engine skips the ~2.3 GB model download, the RAM pressure, and the Neural Engine load entirely — your audio is transcribed fast on our servers instead of your hardware. It's opt-in, clearly labeled in the app, runs only while you have it selected, and we never store your audio on our servers. Included in the same $99/year plan, and you can switch between local and cloud anytime.
Before you start
Permissions & good to know
Accessibility
Lets WhisperJot type the transcribed text into whatever app has focus.
Input Monitoring
Detects your global hotkey (including the fn key) from anywhere on the system.
About the model download
On first dictation, WhisperJot downloads a speech model — Jot Local ≈ 2.3 GB, or a smaller Jot Local Pro model — to your device. It's a one-time download; afterwards local dictation runs offline and launches instantly. On macOS the app lives in your menu bar (no Dock icon), and the download progress shows right next to its icon.
Built on open-source speech research: NVIDIA Parakeet (CC-BY-4.0), OpenAI Whisper (MIT), Silero VAD (MIT).
Optional local LLM
Extra text polish via MLX-LM or Ollama runs on 127.0.0.1.
Plan for 16 GB+ RAM if you enable it.
Specs check out?
Download WhisperJot