Make a short
Find the good bit, reframe it upright, done.
Your file
Drop a file here, or click to choose
It stays on this device. Nothing is sent anywhere.
Settings
Choose the pieces to keep
The two jobs, in one pass
Turning a long recording into something postable is really two jobs: finding the thirty seconds worth keeping, and making a wide video work in an upright frame. This does both at once, so the file is decoded once instead of exported, reopened and exported again.
Find the piece the way that suits the footage. Mark it by hand on the timeline, let it cut out the dead air, ask it for the loudest moments, or teach it a phrase you say when something happens and let it hunt for that. Several pieces become one video with hard cuts between them, which is how most shorts are paced anyway.
Filling an upright frame
A 16:9 video does not fit a 9:16 hole, so something has to give.
- A blurred copy behind is what most shorts use. The whole picture stays visible, and the gaps are filled with an out-of-focus enlargement of the same frame. Nothing is lost and it does not look like an accident.
- Crop in fills the frame properly, at the cost of both sides. Good when the subject is centerd, bad when the thing that matters is off to one edge, so there is a control for which part to keep.
- Plain bars when the whole width matters more than filling the screen.
Loudness
Phone and headset audio is usually well below what a feed expects, and a quiet short gets scrolled past. Evening out the loudness brings it up to roughly the level platforms normalise to, which is the single cheapest thing you can do to make a clip feel professional.
What it does not do yet
Captions are a separate step. Run the short through auto-captions afterwards to burn word-timed subtitles onto it, since most shorts are watched with the sound off. Music, stickers and transitions are not here either; this is the boring part done quickly, not an editor.
Common questions
How long can the source video be?
As long as you like. The file is read from your disk rather than loaded into the tab, so multi-gigabyte stream recordings are fine. Only the piece you keep is processed.
Why is this slower than cutting a video?
Because it is a real re-encode. Cutting copies the original stream untouched; reframing has to decode every frame, rebuild it at 1080 by 1920 and encode it again. Expect roughly a few times the length of the clip on an ordinary laptop.
Is there a watermark?
No. It is plain FFmpeg, so nothing is added to the picture and nothing is added to the file.
Can I add captions?
Not in this tool yet. Make the short here, then run it through auto-captions, which burns word-timed subtitles in the style most shorts use.
Which piece finder should I use?
Cutting out the quiet parts suits a talking recording with pauses. The loudest moments suits gameplay. The phrase finder is best if you say something consistent when a moment happens, since it hunts for those words in your own voice.