Fact · Dictation software
Yes
Whisper Dictate says its audio is transcribed locally by NVIDIA Nemotron models on your GPU, and that nothing is uploaded.
Quote found word for word on the live source, last checked October 22, 2026
In the source's own words
“Audio is transcribed locally by NVIDIA Nemotron streaming models on your GPU (Apple GPU on macOS). Nothing is uploaded”
Whisper Dictate: Offline voice dictation for developers, Whisper Dictate, 2026, Private by architecture section.
Open the original- What it measures
- The maker's home page, in the Private by architecture section.
- What it does not show
- The privacy policy adds that the app also runs on the CPU. The speech models are downloaded once from Hugging Face.
In the data
A row of Dictation App Features Compared: Offline, Privacy, AI, as Whisper Dictate: speech on device: 1 1 = yes, 0 = no.
- App
- Whisper Dictate
- Feature
- Speech processed on the device
- Detail
- runs on your GPU or CPU
- As of
- 2026-10
- Note
- 1 means the maker's page says the app does this