Publishing every Monday grew listeners 10x
Source 04:11 – 05:03 · full clip 0:55
Demo
One long podcast episode in, ready-to-post vertical clips out
These clips were generated by Cutly from a public long-form video. Watch them, then jump to the exact moment in the original to see what the AI picked and how it reframed it.
Steven Bartlett — How I Make $1.2 Million A Year From This PodcastSource video on YouTubeThe clips
10 million views each, yet nobody remembered
Source 22:38 – 23:28 · full clip 0:53
What Cutly did
- 01
Transcribe
The source audio is transcribed with word-level timestamps.
- 02
Select moments
An LLM reads the transcript in overlapping windows, proposes self-contained segments, then ranks them against each other on hook, engagement, value, and shareability.
- 03
Write a hook
Each clip gets a short hook title, burned into the top of the frame for the first few seconds.
- 04
Reframe to 9:16
Each shot is classified. Talking shots get a crop that tracks the speaker's face. Charts, slides, and screens, like the listener graph later in the first clip, show the full frame over a blurred background. Switches line up with the source's scene cuts.
Captions are timed word by word, and the active word is highlighted. The same pipeline runs in the hosted app and in a self-hosted install; the AI video clipper guide covers each stage.
Run it on your own video
Paste a YouTube, Twitch, or Kick link or upload a file. Review the suggested clips, adjust the captions, and export for Shorts, Reels, or TikTok.
Continue exploring
AI Video Clipping
AI video clipper for Shorts, Reels, and TikTok
How Cutly finds moments, scores candidates, reframes faces, and adds captions.
ReadYouTube to Shorts
Repurpose long YouTube videos into Shorts
A practical workflow for selecting, captioning, reframing, and reviewing clips.
ReadSelf-Hosting Guide
Self-host Cutly with Docker Compose
Requirements, environment variables, first boot, a fully local setup, and production checks.
Read