Turn a long video into vertical shorts with AI

It transcribes, picks the moments with the strongest hook, reframes to 9:16 following the face and burns in word-by-word captions. Ready for TikTok, Reels and YouTube Shorts in one step.

moments picked by AI9:16 following the faceword-by-word captions
Try it free ↓ free during beta
long video
9:16 shorts
long video 9:16 shorts · Airpic
JPG · PNG · WebP · TIFF · HEIC · AVIF
How it works

How it works

01

Upload the long video

Podcast, talk, interview, lesson… up to 2 hours and 1 GB. Choose how many clips and how long.

02

The AI picks and reframes

Whisper transcribes, a language model selects the hooks and the 9:16 crop follows the speaker's face.

03

Download the shorts

MP4 1080×1920 with burned-in captions and title, plus each clip's SRT. Deleted in 1 hour.

FAQ

Frequently asked questions

How does it choose the moments?

A language model reads the whole transcript and looks for hooks, stories, surprising facts and advice that stand on their own, always cutting on complete sentences. You choose how many clips and how long.

What about framing?

We detect the speaker's face at each moment and the vertical crop follows it smoothly, without jerks. If there are no faces, it centres the picture.

Can I remove the captions or the title?

Yes, each has its own checkbox. Captions are in the video's language with the active word highlighted, and the SRT comes separately if you prefer the platform's own.

How long does it take?

About 2-4 minutes for a 30-minute video and 3 clips (transcription, selection and rendering), plus 1-2 minutes if the engine starts cold.

Which input formats?

MP4, MOV, WebM, MKV and AVI. Output is MP4 H.264 at 1080×1920 and 30 fps, accepted by every platform.

Do you keep my videos?

No. The video is processed and deleted: 1 hour after download, 24 h at most. No copies.