Glossary
What is split screen?
By the KeyMo team · Updated
Split screen is a layout that shows two or more separate video images side by side, or one above the other, in a single frame. In film it can show two places at once. In podcast clips it usually puts each speaker in their own box, and in 9:16 the boxes are stacked.
Split screen meaning
Split screen puts two or more shots in one picture, each in its own region, and every region plays at the same time. In film it can show two places, two people on a phone call, or two moments at once. In everyday computing the same words also describe running two apps next to each other on a phone or laptop. This page covers the video layout.
Split screen for a two-person podcast
If you record a conversation on two cameras, or a video call, you have two pictures. A split screen keeps both on screen so a viewer sees the speaker and the listener's reaction. In a wide 16:9 frame the two boxes sit side by side and each is still a roomy shape. A 9:16 frame is the problem.
| Layout on 1080x1920 | Size of each box | What you crop from a 1920x1080 source |
|---|---|---|
| Stacked, top and bottom | 1080x960 each | About 1215 pixels wide by the full 1080 high, so roughly 63% of the width |
| Side by side | 540x1920 each | About 304 pixels wide by the full 1080 high, then scaled up about 1.8 times |
Both rows use the full height of a 1080-pixel source. Stacked boxes are close to square and hold a face well. Side-by-side boxes are slivers: a face fits, a shoulder does not, and the upscale softens the picture. That is why vertical podcast clips nearly always stack the two speakers.
Split screen or single frame?
- Two people, one moment of overlap or a reaction that matters: stack them.
- One person talks for most of the clip: stay on one speaker and cut to the other when they reply.
- Different rooms or a video call: stacking keeps the call feel and avoids cropping one face into a bad shape.
- Both in one wide shot at a table: crop to whoever speaks, since splitting a single wide shot just cuts the table in half.
Worked example
A 1920x1080 recording has host and guest on separate cameras. For a 40-second clip, crop each camera to about 1215x1080 around the face, scale it to 1080x960, and stack the guest on top of the host. Keep the captions across the seam between the boxes, away from the bottom edge where the app draws its own buttons. See safe zone.
Common mistakes
- Squeezing two faces side by side into a 9:16 frame so each is a thin strip.
- Using a split when only one person talks, which halves the size of the face for no gain.
- Captions that cross both boxes at the same height as a chin or forehead.
- Forgetting that each box needs its own head position, because two faces at different sizes look uneven.
Nearby terms and KeyMo
Split screen is different from picture-in-picture, where a small video sits over a large one. A multicam recording is how you get the separate shots, and a two-shot is the opposite approach, one frame holding both people. KeyMo takes a different route to the same two-person problem: it reframes to one 9:16 frame and follows a face with automatic face tracking.