Cloud video search means permanently hosting your footage on someone else's servers. Local search keeps your index and your originals on your machine.
Frames and audio are extracted on your device and sent for analysis, nothing else. Your originals never leave your machine, and the extracted data is discarded the moment analysis completes. Your search index stays on your machine.
Millisecond results from a local index, with no round-trip to a server.
Frames and audio are extracted on your device and sent for analysis. Your full-res originals stay put, and the extracted data is discarded the moment analysis completes.
Search on planes, in remote locations, or anywhere without internet.
Full multimodal video search
FrameQuery indexes your entire video library with transcription, object detection, face recognition, and scene detection. Search by what people say, what appears on screen, or the mood of a scene. Supports 50+ formats including RED R3D, ProRes, and BRAW.
Desktop media browser
Kyno is a desktop media browser for previewing and organising footage. It searches metadata fields (filename, tags, descriptions) and does not search inside video content.
PC-only video cataloguing
Fast Video Cataloger generates thumbnail grids and lets you tag clips manually. It is Windows-only and does not include AI-powered search. Search covers only the metadata you add yourself.
On-premise transcription
Simon Says offers an on-premise deployment for organisations that need transcription without cloud uploads. It handles transcript search well but does not support visual search (objects, faces, scenes). The price point puts it out of reach for most freelancers and small teams.
| Feature | FrameQuery | Kyno | Fast Video Cat. | Simon Says |
|---|---|---|---|---|
| Transcript search | ||||
| Object detection | ||||
| Face recognition | ||||
| Scene detection | ||||
| 50+ formats | ||||
| Works offline | ||||
| macOS + Windows | ||||
| NLE export |
Object detection, transcription, and scene description are GPU-heavy. Most video search tools run them in the cloud.
FrameQuery runs that processing in the cloud during indexing and stores the resulting search index on your machine. Face and voice recognition run on-device. Once indexed, every search is local and offline.