Smart Recognition
Tag once, find everywhere
Name a face in one video and FrameQuery finds that specific person across every clip in your library. New imports are matched automatically - no re-tagging required.
People
Sarah
3F 1V
James
5F 2V
Faces
3
Right-click or drag onto a person to assign
Voices
2Right-click or drag onto a person to assign
All biometric data processed and stored on-device only
Privacy First
Your biometric data stays on your device
Face and voice recognition are optional, consent-gated features. Models download to your machine and run locally with GPU acceleration. Embeddings are encrypted at rest and never uploaded. Biometric data is excluded from shared indexes entirely.
On-device processing
Face and voice recognition models run locally with GPU acceleration. No biometric data is ever sent to our servers.
Encrypted storage
Face embeddings and thumbnails are encrypted at rest. Only accessible when the app is open and you are signed into the account that created them.
Consent-gated
Nothing runs until you explicitly opt in with two separate consent checkboxes. Full audit trail of consent events.
Compliant by design
BIPA, GDPR, and CCPA compliant. Biometric data is excluded from shared indexes. A dedicated “Delete all biometric data” action cascades removal from every video and person in your library.
People Management
One name, every appearance
Once you name a person, FrameQuery links that identity across your entire library. New videos with matching faces are tagged automatically. Merge duplicates, rename, or delete people and changes propagate instantly.
Voice Recognition
Faces and voices, linked automatically
Voice recognition works alongside face detection. Speaker embeddings (ECAPA-TDNN) identify unique voices from the audio track. If a face and voice appear in the same frame, FrameQuery links them to the same person automatically. Filter transcript results by speaker.
Available on every plan including Free. Enable when you're ready and FrameQuery backfills any previously indexed videos.