Publishing every Monday grew listeners 10x
Source 04:11 – 05:03 · full clip 0:55
These clips were generated by SupoClip 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 YouTubeSource 04:11 – 05:03 · full clip 0:55
Source 22:38 – 23:28 · full clip 0:53
The source audio is transcribed with word-level timestamps.
An LLM reads the transcript in overlapping windows, proposes self-contained segments, then ranks them against each other on hook, engagement, value, and shareability.
Each clip gets a short hook title, burned into the top of the frame for the first few seconds.
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.
Paste a YouTube link or upload a file. Review the suggested clips, adjust the captions, and export for Shorts, Reels, or TikTok.
AI Video Clipping
How SupoClip finds moments, scores candidates, reframes faces, and adds captions.
ReadYouTube to Shorts
A practical workflow for selecting, captioning, reframing, and reviewing clips.
ReadSelf-Hosting Guide
Requirements, environment variables, first boot, a fully local setup, and production checks.
Read