Fact · Languages and accents
Voice Key's privacy page says that recognition runs on the user's device in both its classic and its streaming mode.
Quote found word for word on the live source, last checked October 25, 2026
In the source's own words
“Classic and Streaming modes both run locally.”
Voice Key privacy page, BuildWithAIs, 2026, Local speech recognition.
Open the original- What it measures
- The privacy page describes how Voice Key processes audio. Local recognition uses downloaded models. Cloud refinement and translation are separate, optional steps and are off by default.
- What it does not show
- This is the maker's description of its own app. The page also says that model downloads and update checks use the network, so the app is not offline by default. The claim covers recognition, not the optional AI refinement.