Where Deixis runs
Deixis is one engine with four hosts: Windows, Mac, iPhone and Android. Speech recognition runs on the device on every one of them, and the cloud is only ever for the optional cleanup, translation and meeting notes. This page is the platform table from the documentation: what each host does today, and which pieces are built but not yet released.
Where Deixis runs
One engine, four hosts. Speech recognition runs on the device on every one of them, and the cloud is only ever for the optional cleanup, translation and meeting notes, exactly as on Windows.
| Platform | What runs there | State, 26 September 2026 |
|---|---|---|
| Windows 10 and 11, 64-bit, AVX2 | Everything in this README. | Shipping: the signed installer and the in-app updater. |
| macOS, Apple Silicon, 13 and later | Hold-to-dictate at the caret (the chord is Ctrl+Option rather than Ctrl+Win), with cleanup sent the same fields under the same privacy rules; the settings, history and dictionary windows; free mode and the toolbar, screen grab and annotation, screen recording. |
Shipping: a notarized disk image and the in-app updater. |
| iPhone | A keyboard: switch to it, hold the microphone key, speak, and the text lands in the field. Whisper runs inside the Deixis app on the phone; the keyboard only inserts. | In beta on TestFlight; not yet in the App Store. Meetings on the phone are not there yet. |
| Android | A floating microphone bubble that dictates into any app's text field, and meetings recorded from the phone's microphone, transcribed on the phone as they run, with notes. Translate, the Library, the dictionary, the account and both speech models carried over. | Shipping: on Google Play. |
Landing in 0.9 for meetings, on the desktop: name the people in a meeting instead of "Me" and "Them", and voice identification that tells the remote voices apart and, once you have named a voice, recognises that person in later meetings. The voiceprint is a mathematical fingerprint that stays on the machine, is never uploaded, and can be deleted per person.. Until it lands, the Meeting mode section below describes the "Me" and "Them" labelling as it ships today.
The architecture of each port, and the work still open on it, is in MACOS.md, IOS.md,
ANDROID.md and CAPTURE.md; the tracker carries the tickets. Those belong beside the code rather
than in this table, which is read by people asking one question: can I run it?
Keep reading
From the blog
Speech to text offline: how to keep dictating when the internet goes
How to dictate with no internet on Windows, Mac, Android and iPhone, what offline really covers, what to do before you go, and what still needs a network.
How to type by voice in any Android app, with the speech recognised on the phone
Dictate into any Android app's text field from a floating microphone. Speech recognition runs on the phone, and meetings are transcribed there too.
How to type by voice in any app on a Mac, with the speech recognised on the Mac itself
How to type by voice in any app on a Mac: hold Ctrl+Option, speak, let go. Speech recognition runs on your Mac, free and with no account.
Hold a key. Speak. Keep working.
Deixis is free, needs no account, and runs on your own device.
v1.2.2 · 64‑bit Windows 10/11 · macOS 13 on Apple silicon · 51 MB · nothing else to install · also for Windows