VelvetSpeak

Privacy

VelvetSpeak listens with you during a conversation and helps in the moment. Here's exactly what that involves: what's handled live, what's kept, and what you control. This is a small closed beta, so it's short and specific rather than a legal document.

What's processed live

Audio from your glasses becomes a transcript on your device or phone. That transcript is how the lens understands the conversation.

To generate a suggestion, the app sends what it needs to the VelvetSpeak backend and gets an answer back. Those requests are for producing the next card, not a record we keep.

What's stored

  • Memories, according to the consent level you pick: Minimal, Standard, or Full. Higher levels let the app remember more across sessions; Minimal keeps almost nothing.
  • No transcripts server-side. Transcript text stays on your phone. We don't store the words from your conversations on the server.
  • Content-free usage metering: counts and durations only (how many sessions, how long, which lens). This keeps the beta within its limits. It never includes what was said.
  • A VoiceLock profile, only if you turn it on. VoiceLock is optional and off by default. If you set it up, it uses short samples of your own voice to build a recognition profile that tells your voice apart from others. We keep that profile, not the recordings — the raw enrollment audio is discarded — and you can delete the profile anytime in Settings or with Wipe all.

Your controls

  • Export: download what's stored about you as JSON.
  • Wipe all: delete your stored memories.
  • Revoke your key: turn off access for a device.
  • Consent level: change Minimal / Standard / Full at any time; it applies going forward.

Questions or requests

This is an invite-only closed beta. If you have a privacy question or want your data exported or wiped, contact the person who invited you. They can reach the team directly.