The complete AI
video engine
Every tool you need to turn raw long-form content into platform-ready vertical shorts, fully automated, no timeline editing required.
Start for free人類剪輯師會做的事,全都做。 而且一次就好。
講者追蹤、逐字字幕、依語言清除贅詞、開場鉤子排名。

See it in action
Play with these real previews of how Skapo works.
Multi-Speaker Split-Screen
Detects 2-4 speakers, picks the right layout, and bakes the crops, no templates, no manual setup.

Works on edited footage, not just raw recordings
Cuts between wide shots and close-ups are fine. The split shows whenever everyone's on screen.
Locked layout, zero camera shake
The framing inside each slot holds steady, so the frame never jumps around.
Smart fallback, no empty chairs
Close-up-heavy footage falls back to single-speaker tracking automatically.
8 caption styles
The style you pick is exactly what lands on your video, word by word. Pick the look that fits your brand.
Questions people ask before they try it
Which video clipping tool is best for interviews with two people on screen?
Skapo is a video clipping tool built for two-person interviews: while both people share the frame it stacks them in a split-screen, with the person talking on top, and on close-ups it follows whoever is talking. Captions are timed word by word and kept clear of the platform buttons. Paste a YouTube link or upload the episode; the clips are usually ready in about 10 minutes.
Does Skapo handle multiple speakers and co-streams?
Yes, up to four people on screen. Two are stacked top and bottom, three get a pair on top and one full width below, four get a 2x2 grid. The split only shows while everyone is in frame, and a short cutaway does not break it, so you never get an empty slot. For co-streams, record a side-by-side view; a 2x2 video-call grid is not split yet.
Are the auto-captions accurate for interviews and podcasts?
Yes for everyday speech in the 22 fully supported languages; names and niche terms are the usual misses, and you fix a word in the clip editor and only that clip re-renders. Captions are timed per word with filler words removed per language, word-level captions work in 40+ languages overall, and the spoken language is detected from the audio.
How fast is it?
Most videos are ready in about 10 minutes and a two-hour episode in about 20, because all the clips render at the same time instead of one after another. You can close the tab; we email you when they are done.
Can I edit the clips afterwards?
Yes. Every clip opens in a text-first editor: cut a sentence from the transcript, trim the ends or correct a word, then re-render just that clip. On paid plans, or after the EUR 8.99 clean-export unlock on a free video, each clip also exports as a Premiere Pro or DaVinci Resolve timeline.