J-cut / L-cut
A split edit where sound and picture change at different moments, so audio from the next shot arrives early (J) or the old one lingers (L).
See it
What it is
The names come from the shape the clips make in a timeline. In a J-cut the incoming shot's audio starts before its picture, so you hear the new scene first. In an L-cut the outgoing shot's audio keeps running under the new picture. Both are 'split edits': the cut point for sound and the cut point for image are not stacked on the same frame.
Reach for them in dialogue and interviews. Cutting sound and picture on the same frame lands like a door slamming, and the viewer feels the edit. Offset by half a second to two seconds and the edit disappears while the ear pulls you into the next shot. Split edits are extremely common in dialogue scenes and interviews, and documentary editors lean on them to slide under B-roll.
Gotcha: split edits need room tone. If two clips have different background noise floors, the overlap becomes an audible hiss swap, louder than the cut you were trying to hide. Also unlink audio from video first, or dragging one just drags both.
Ask AI for it
Edit this dialogue sequence with split edits. At each dialogue cut, choose one: a J-cut when the incoming line should lead us into the next shot, starting that audio 0.5 to 1.5 seconds before its picture, or an L-cut when the outgoing line should keep running about 1 second over a reaction shot. Do not apply both at the same cut unless the dialogue is meant to overlap. Unlink audio from video, place the picture cuts on natural gestures rather than on sentence ends, and crossfade room tone by 6 to 10 frames at every audio boundary so the noise floor never jumps.