Fact · Voice privacy and security
VivaDicta's privacy policy says that when you use on-device models, your audio and text never leave your device, and that it has no access to your recordings or transcriptions.
Quote found word for word on the live source, last checked October 22, 2026
In the source's own words
“When using on-device models, your audio and text never leave your device. All processing happens locally, and we have no access to your recordings, transcriptions, or AI-processed text.”
VivaDicta Privacy Policy, VivaDicta, VivaDicta, 2026, On-Device Processing section.
Open the original- What it measures
- The section lists the on-device transcription choices (Whisper, Parakeet and Apple Speech) and the on-device AI choices (Ollama and Apple Foundation Model). Cloud services are a separate, optional section.
- What it does not show
- This covers on-device models only. When you choose a cloud provider, your audio or text is sent to that provider and its own policy applies. The policy gives no number of days.
In the data
A row of How Long Voice Assistants Keep Your Recordings, as VivaDicta: on-device, no access: When using on-device models, your audio and text never leave your device. All processing happens locally, and we have no access to your recordings, transcriptions, or AI-processed text. days.
- Service
- VivaDicta
- What is kept
- Audio and transcripts
- When
- Other
- Reviewed by people
- No
- As of
- 2026-10