Clipto - Fully local, natural language search over terabytes of media
by•
Like Google Photos, but fully local. Turn the terabytes of video, audio, meetings, and files you work with into searchable memories, without uploading anything to the cloud. Clipto automatically tags people, dialogue, and scenes, so you can instantly find any moment buried in your media just by describing what you're looking for. It's fast too: on a MacBook Pro M5, Clipto indexed 2TB of videos in just 24 hours.
Replies
Lessie AI
Can I drag and drop clips directly from the Clipto search window straight into my Premiere Pro or DaVinci Resolve timeline, or do I need to reveal in Finder first?
Clipto
@libin_yao Yes. In fact, we’ve already built a Premiere Pro plugin specifically for this workflow.
You can search your media directly inside Premiere using Clipto, find the exact moment you’re looking for, and add the selected clip to your timeline without jumping back and forth between Finder and your editor.
For many editors, the goal isn’t just finding the clip, it’s finding it without breaking creative flow. That’s one of the main reasons we built the integration in the first place.
If you have to choose, which one you use more heavily? Premiere or Davinci?
"Automatically tags people" — is that face recognition, voice matching, or something else? And when it misidentifies someone, is there a way to correct the label without re-indexing the entire library?
Clipto
@sounak_bhattacharya Yes! We actually use both visual face recognition and voice identification to build a more complete understanding of who appears across your media.
And yes, corrections are fully supported. If Clipto misidentifies someone, you can simply relabel that person (or merge/split identities), and the change is reflected throughout your library. There’s no need to re-process or re-index everything from scratch.
In fact, user corrections become part of the local memory layer, which helps make future search and retrieval much more accurate for your own media collection.
This is pretty compelling for on-prem AI model training, especially in regulated industries. We've been hesitant to use cloud-based media analysis for compliance reasons, and this solves that. I can see our data science team using this to curate datasets from internal meeting recordings without needing a dedicated labeling pipeline.
Clipto
@kobejordan That’s a really interesting use case.
We originally built Clipto around search and retrieval, but dataset curation is a natural extension, especially when recordings can’t leave a secure environment.
We’ve heard similar concerns from teams in legal, healthcare, and other privacy-sensitive domains where cloud workflows simply aren’t an option.
Would love to learn more about the kinds of data your team is working with.