Skip to content

Fact · Dictation software

11 = yes, 0 = no

Freeway's privacy policy says it processes all speech recognition entirely on the user's device, using local neural engines.

Checked against its source on October 26, 2026

In the source's own words

“Freeway processes all speech recognition entirely on your device using local neural engines.”

Privacy Policy | Freeway, Synthilab OÜ, Synthilab OÜ, 2026, Our Core Privacy Principle.

Open the original
What it measures
The app uses NVIDIA Parakeet, optimized for Apple Silicon through CoreML, per the homepage.
What it does not show
This describes the app by default. Plugins are third-party components outside the policy, and a plugin that calls an LLM would send text out (see the sends-text row).

In the data

A row of Dictation App Features Compared: Offline, Privacy, AI, as Freeway, speech on the device: 1 1 = yes, 0 = no.

App
Freeway
Feature
Speech processed on the device
Detail
all recognition on the Mac
As of
2026-10
Note
1 means the maker's page says the app does this.

Source