The picture and sound do not have to change at the same frame. An L-cut can continue the outgoing speaker's audio over the next picture, while a J-cut can introduce incoming audio before its picture appears. Choose the relationship that preserves the sentence and intended reaction. Moving both boundaries together can accidentally remove a final word or make a reaction appear to answer a different line.
Write what the viewer should understand
Mark the complete spoken thought and the reaction that belongs with it. Decide whether the audience should hear the current sentence finish while seeing the listener, or hear the next voice before seeing its speaker. These are narrative decisions, not simply ways to make a cut less noticeable.
Adobe documents J-cuts and L-cuts through separate audio and video edit boundaries. Use the application's supported split-edit controls and preserve the linked source relationship where appropriate. Do not delete wanted speech just because the picture no longer shows its speaker.
Build a two-boundary edit card
On a duplicate sequence, identify the proposed picture change and the independent audio change. Check available source material and useful sound on both sides. There must be real recorded content to extend; moving a boundary does not create a missing word or a genuine listener response.
| Decision | Mark in the actual sequence | Check |
|---|---|---|
| Picture switches | Intended reaction begins | Expression belongs to this context |
| Outgoing speech ends | Complete word or thought | No clipped consonant or altered meaning |
| Incoming sound begins | Appropriate acoustic transition | No accidental double speech |
This card is a proposed editorial plan. It does not claim a supplied reaction was recorded at that moment.
Preserve context in a hypothetical interview
Suppose a speaker says a qualification after a strong statement, while the best listener shot starts slightly earlier. Switching picture during the line can work if the full qualification remains audible and the reaction is ethically and contextually appropriate. Cutting the sound at the picture boundary could remove the qualification and change the message.
Check room tone and overlapping voices as well as the words. A split edit may need an appropriate transition to avoid a sudden ambience jump, but excessive crossfading can create doubled syllables. Listen without watching, then watch without relying on captions. If source handles are insufficient, choose another picture boundary or shot rather than implying missing material exists.
Approve meaning and mechanics together
Render the sentence before the transition, the reaction and the next complete thought. Verify no word is lost, no reaction is misleadingly attributed and picture timing remains intentional. Save both edit points in the handoff notes. This differs from deciding whether narration is on-screen or off-screen in a generated film: the task here is a concrete split edit of existing picture and sound with separate boundary evidence.
Sources and scope
Prepared by Imagild Editorial with AI assistance. The example is a proposed diagnostic, not a reported software test. These are external post-production controls; their documentation does not establish an Imagild feature.