Talk to Your Mac and See Your Words Appear, Even Offline

Jexxa is a Mac app that turns your speech into text right on your laptop, with no internet needed. The language model lives on your disk. The current release is version 0.1.10 as of September 15, 2026 jexxa.org.

Compatibility and privacy

Jexxa needs macOS 14 or later and a Mac with Apple silicon. There is no Intel version listed. Building for one chip family makes it simpler to run fast on Apple's AI hardware and shared memory design, without the slowdowns of older Macs.

The website states that no audio or transcript is uploaded from the Mac, and that the software works offline without Wi-Fi. Your speech is decoded on the laptop itself, so there is no wait for a server reply and private speech does not travel over the internet or sit in outside logs.

Speed and voice editing

Jexxa's website states that most dictations are typed within a fifth of a second of releasing the dictation key. You also see a live preview of the text while you speak. The preview helps you catch mistakes early, and the fast finish keeps dictation usable inside editors, terminals, browsers, and other text fields without breaking your flow.

You can also fix text by voice. Saying "JX minus one," "JX minus two," or "JX clear" deletes the last line, the last two lines, or all dictation. These short fixed phrases are easier for the system to hear correctly than open instructions, and they let you roll back misheard text without using selection keys.

Learning, size and price

Jexxa learns corrected names and jargon after you make a fix, and that learning can be undone. This helps with a common weak spot. Proper nouns, product names, API identifiers, and domain terms are rare words the base model often misses. Learning from a correction fixes that for next time, while undo gives you a way back if a correction keeps causing the same mistake.

It comes in two downloads, like choosing a larger or smaller suitcase for the same trip. Full JEXXA is a 3.1 GB download for Macs with 16 GB or more memory, and JEXXA Small is a 2.0 GB download for 8 GB Macs. That choice matters because the model shares memory with IDEs, browsers, containers, and other programs running at the same time.

Pricing is $8 per month plus tax, with unlimited dictation and no per-minute billing.

What it means in daily use

The broader tradeoff here is using local disk space and memory to get behavior you can count on. Cloud dictation saves local resources but ties everyday typing to the internet connection, server queues, and data-handling policy. A 2.0 to 3.1 GB local package flips that. You spend disk and working memory once, then get steady offline use and results in under a second.

In my view, that is a reasonable exchange for professionals who dictate in source code discussions, customer notes, medical or legal drafting, or any setting where transcripts are sensitive and Wi-Fi is not guaranteed.

The broader context for evaluation is how the learning holds up over time. Short-term adaptation to names and jargon is useful. Long-term value depends on whether learned terms stay scoped correctly, survive updates, and stay easy to inspect and remove. Jexxa's inclusion of an undo control suggests attention to that problem, and hands-on testing with dense technical vocabulary would show how well it holds up.