Is AuraScribe really free?

Yes. AuraScribe is free forever and open source under the MIT license. There is no paid tier, no subscription, and no account to create.

Does my voice get sent to the cloud?

No. Every word is transcribed on your own PC. The only network request the app ever makes is downloading a speech model once, after which it works fully offline.

How is it different from Wispr Flow, Superwhisper, or Dragon?

AuraScribe is free, open source, and 100% offline, with no account required. Wispr Flow and Superwhisper are paid and cloud-based; Dragon is expensive. It is the private, local alternative.

Which languages does it support?

English plus 25 European languages, roughly 40 Asian languages, and Malayalam and Kannada, using four on-device engines. Each auto-detects the language within its region.

What are the system requirements?

A microphone and one of: Windows 10 or 11 (64-bit), macOS on Apple Silicon, or Linux (Debian or Ubuntu). The download is about 9 MB and runs faster than real time on an ordinary CPU, no GPU required. No internet is needed after the first model download.

Is there a macOS or Linux version?

Yes. As of version 2.0, AuraScribe runs on Windows, macOS (Apple Silicon), and Linux (Debian and Ubuntu), all from the releases page. Windows is the most battle-tested; macOS and Linux are newer. On macOS you approve it once in Privacy and Security (it is not notarized) and grant an Accessibility permission; on Linux it works best in an X11 session. The download button picks the right build for your system automatically.

Does it work in every application?

Yes. Press the hotkey in any window and the finished text is typed straight into whatever app has your cursor: browsers, editors, terminals, chat boxes, and form fields.

AuraScribe on Linux: offline dictation for Debian and Ubuntu

Set up AuraScribe, a free open-source offline dictation app, on Debian or Ubuntu: install the .deb, use an X11 session for the global hotkey, and type clean text into any app, fully offline.

Good offline dictation has been hard to find on Linux. AuraScribe runs there natively, does all of its transcription on your own machine, and types clean text into any application. This guide covers installing it on Debian or Ubuntu and the one session detail that matters for the global hotkey.

Install the .deb

Download the Linux package (a .deb) from the releases page and install it with your package manager, for example sudo apt install ./AuraScribe_2.0.0_amd64.deb, or by double-clicking it in your file manager. It targets Debian and Ubuntu and their derivatives.

Use an X11 session

This is the one thing to know on Linux. A global hotkey that works in every window, and the ability to type text into whatever app has focus, both rely on synthetic input, which Wayland deliberately restricts for security. So AuraScribe works best in an X11 login session. On the login screen, click the gear or session icon and choose the Xorg or X11 option before you sign in. Everything else, including the transcription itself, behaves the same as on Windows and macOS.

Set the hotkey and dictate

The default shortcut is Ctrl + Shift + Space, rebindable to anything you prefer. Press it in any window, speak, and release. Recognition runs on your CPU as you talk, and a local cleanup pass handles punctuation, casing, and filler words, so the finished text is typed straight into your editor, browser, terminal, or chat box the moment you stop.

What you need

AuraScribe on Linux needs a 64-bit Debian or Ubuntu system, a microphone, and, for the hotkey and typing, an X11 session. No GPU is required. After the first language model downloads once, it runs fully offline, with no account and no subscription.

If you want dictation that never leaves your machine, that is the whole idea here. You can download AuraScribe for Linux now, or read more about local versus cloud dictation.

All posts