What is ClipVase?
ClipVase is an iPhone app that makes the rough cut from your own camera roll. Paste a script, say a voiceover, or pick a video of you talking; it finds the B-roll for every line and trims each clip to your voice. The cut leaves as clips in order plus your voice as its own track, to Photos, to your editor, or to your AI. Nothing is uploaded.
Does it use stock footage, or my own videos?
Only yours. ClipVase never shows you a clip you didn't shoot. If you want stock B-roll there are plenty of tools for that; this one is for the footage already on your phone.
How do I find a video I know I filmed but can't find in my camera roll?
Photos search finds one thing at a time: type "beach", get beach videos. ClipVase reads your whole script and shows you matching clips for every line, in order, trimmed to your voice. Search the way you remember it, like "the kitchen at 5 AM", and it looks for that in what each clip shows.
Can I make a video from my camera roll without writing a script first?
Yes. Open Footage, tap a folder like cooking or beach days, pick a few clips and say one voiceover over them. ClipVase trims each clip to its part of what you said. The script comes out of your mouth, not a blank page.
How do I get my clips to match my voiceover without trimming each one by hand?
ClipVase hears where your words fall and gives each picked clip its share of the voiceover, so the clips change when your sentences do. Drag a clip to start it somewhere else, or change its speed, and the length still fits your words.
How do I add B-roll over a video of me talking?
Yes. Pick the video and ClipVase splits what you said into lines. Keep yourself on camera for a line, or show a clip from your footage over it. Your voice runs underneath the whole time.
Does it make the video for me?
No. It finds and trims. You get your own clips, picked and in order, plus your voice as its own file: a rough cut, ready for CapCut, Final Cut, or your AI to finish. The video stays yours to make.
Is my footage uploaded to the cloud?
No. Sorting, search, speech to text and export all run on your iPhone. There is no account and no ClipVase server. Two things leave the phone, both by your choice: a post you send to your AI app, by AirDrop, to Photos or to an app you pick; and, to name Where folders, a rounded location to Apple's own place-name service, never a video.
Do I need to tag or organize my videos first?
No. Folders are made for you from what is in each clip. Keep a name, use your own words, or tell ClipVase a group is not a group. Your Photos library is never changed.
I have thousands of videos on my phone. Will it still work?
Yes, and you don't wait for it to finish. It starts with your newest videos and works back. Search and folders work from the first clips it has seen, and it tells you plainly how far back it can see.
Does it work with CapCut or my editor?
Yes. Your picked clips and your voice land in one Photos album, in order, with the voice on its own track. Or share the set straight to another app.
Does it work with my AI agent?
Yes. When your post is ready, pick With an AI app on this iPhone: the share sheet opens, your clips, your voice and the script attach on their own, and a note telling the AI what the files are is already in the message box. Tap Send. Using an AI on your Mac instead? Pick With AI on your Mac: one tap AirDrops the same post as a folder, and your AI reads the note inside it.
Can my AI agent see my whole camera roll?
No. ClipVase only hands your AI what you share: the posts you send, one at a time. Nothing is shared until you tap, and the app never reads your Hidden album. What your AI can see elsewhere is between you and that app.
Do I need an AI agent?
No. The same tap sends the cut to CapCut, Final Cut or Photos. The AI door is there for the people who already edit with one.
Which iPhones does it run on?
iPhones on iOS 26. On iPhones with Apple Intelligence, folder names are more specific, like "making pasta". Other iPhones get plainer names, like "food", that you can rename in a tap. Speaking a voiceover uses Apple's on-device speech recognition.
ClipVase
Join the waitlist