Phone clips, camera files, a screen recording — drop in what you shot. Then fill a short form: what the video is, who it's for, which languages you need.
No timeline, no tracks, no project settings. That form is the whole setup.
The AI analyzes your footage before it writes a word — what happens, where the moments are, what the recap should be about.
You don't mark anything up. It does the watching so you don't have to.
00:58 —Script. It writes the narration from what it saw: a recap script in each language you chose, in plain words you can read and change.
01:40 —Voice. A generated voice reads that script, timed to your footage. You never record a line.
Subtitles are drawn into the frame itself, timed to the voice, so they show up the same everywhere the video plays — no caption setting to switch on, nothing a platform can drop.
You still get the subtitles as a separate SRT file in the export bundle, for platforms that prefer their own captions.
Before anything exports, you see it laid out in one view. Three columns, left to right — the same order the work happened in.
Your footage, as uploaded. Check the moments it picked.
The narration it wrote. Change a word, and the voice follows.
The finished recap, subtitled, in each language. Export when it's right.
Every export comes as a bundle. Take any piece of it anywhere.
