Fact · Dictation software
TalkingKey's privacy policy says audio is processed locally by the on-device Whisper engine on the user's Mac.
Checked against its source on October 26, 2026
In the source's own words
“Audio is captured only while you hold the push-to-talk key. It is processed locally by the on-device Whisper engine (whisper.cpp) and used solely to produce your text.”
Privacy Policy: TalkingKey, TalkingKey (PT ITropical Live Solutions), 2026, Section 3, Your voice and audio.
Open the original- What it measures
- The policy's description of how dictated audio is handled.
- What it does not show
- The policy covers the transcription step. The model is downloaded once from Hugging Face, which the policy also discloses.
In the data
A row of Dictation App Features Compared: Offline, Privacy, AI, as TalkingKey: speech on device: 1 1 = yes, 0 = no.
- App
- TalkingKey
- Feature
- Speech processed on the device
- Detail
- on-device Whisper engine (whisper.cpp)
- As of
- 2026-10
- Note
- 1 means the maker's page says the app does this