One long video in, ten clips out
Drop in a podcast, interview, or stream. AI finds the moments worth posting, crops them vertical around whoever is speaking, and burns in captions — ready for TikTok, Reels, and Shorts.
How it works
Drop in a long video
Paste a YouTube link or upload a file up to three hours. Only want part of it? Pick a timeframe and everything after works from that window.
AI finds the moments
Your audio is transcribed with word-level timing, then read for self-contained moments — a hook, a story, a claim that lands. Each clip comes back with a title, a score, and why it was picked.
Post-ready clips come back
Every clip is cropped vertical, captioned, and rendered. Open any of them in the editor to cut, re-word, or restyle before you post.
Built to make clips worth posting
Cuts on complete sentences
Clips open and close on a finished thought instead of a word count. A clip is allowed to run slightly long to finish a sentence, because a truncated line is worse than a few extra seconds.
Framing that follows the speaker
A clip usually spans several camera cuts. Each shot is framed on whoever is talking, picked by mouth movement rather than by who is biggest in frame — so a wide two-shot does not leave your speaker at the edge.
Word-timed captions
Captions are timed per word, not per line, and burned into the video so they survive every platform. Five styles and four fonts, with the position under your control.
A real editor, not a preview
Trim, split, and delete sections. Rewrite or remove any word. Drop in b-roll, images, and text overlays. Re-export as many times as you like — editing never costs credits.
Pick a caption style
Five styles, four fonts. Change them per clip in the editor.
Change anything before you post
Every clip opens in a full editor. Cut a section out, fix a word the transcription got wrong, move the captions, drop in b-roll, or switch the style entirely — then export again. Editing and re-exporting never cost credits.
Try the editor