Section 6: Lip Sync and Performance · Lesson 4 of 5 (29 of 45) · 60 minutes · Evidence: Revised dialogue edit plan
Feature films rarely show a speaker’s face for an entire speech. The camera cuts to the listener, to a hand on a doorknob, to the rain on a window. Editors do this for rhythm, and also because it protects the performance. Today you will learn to use the same tools to cover the weak seconds your generator cannot fix.
Production journal: write your answer below.
What must a filmmaker understand before they can use shorter lines reaction shots cutaways framing and sound bridges to protect performance?
No generator gives a flawless take every time. The professional answer is not to give up on the shot or to pretend the flaw is not there. It is to design the edit so the audience sees the strong frames and the story still makes sense. This is honest craft: you choose what to show, and you never lie about what a performance is.
Coverage is the set of different shots you have of a scene, so the editor has choices. For a dialogue scene you might plan a wide establishing shot, the speaker, the listener, and two or three cutaways. Planning coverage before you edit means you can swap in a shot when a take fails.
A student’s take of Tomas shouting over wind is good for the first four seconds and then the jaw warps. The student does not cut to black. They trim Tomas to the good four seconds, then cut to Mara’s face, glasses fogged by sea mist, for her reaction while Tomas’s voice carries on over her shot. A sound bridge keeps the line unbroken. Next comes a cutaway of the boat’s rope slapping the dock. The audio is unchanged; the picture now avoids the flaw. The student marks the cut points in a CapCut timeline and notes in the plan which shot hides which defect.
Editors working with performance footage are quiet problem solvers. When a take is flawed they do not regenerate forever; they look at the timeline and ask what picture can sit on top of the audio. Study how they plan this, and decide at each step what the tool does and what the editor decides.
First the editor marks the moments in the dialogue track where the performance is weak and where it is strong. Then they list what other shots they have: a listener, a hand, a location, an object. They place the weak seconds under a different picture and slide the cut a few frames either way until the sound feels natural. Finally they play the whole scene without stopping and check that the cut points do not draw attention to themselves.
A tool can place clips and keep them in sync, but it cannot tell you whether the cutaway means something. That depends on the story.
As you study, record each decision point in your production journal.
| Decision point | What the filmmaker does | What the tool automates | What still needs human review |
|---|---|---|---|
| Mark strong and weak moments | Notes times where the performance holds or fails | Nothing; this is review | That weak moments are marked honestly, not only the obvious ones |
| List available coverage | Inventories listener, object, and location shots | Can generate new supporting shots if requested | Whether each shot is authorized and fits the story |
| Place cutaways and reactions | Chooses which picture covers which line | Places and snaps clips on the timeline | Whether the cutaway preserves the meaning of the line |
| Smooth the sound | Adds a sound bridge and shifts cut points | Aligns clips and shows the waveform | Whether the cut feels natural at normal speed |
Studio task: Rebuild a weak talking-head scene using coverage and sound continuity.
Your teacher provides a weak talking-head scene: one long take of a speaker with a flawed last third and no other footage. Work in a small group to rebuild it using coverage.
Coaching tips. Read the line aloud and ask what the audience should see while hearing it. If the answer is “the speaker”, cut away later. If you need the audience to feel the listener’s reaction, cut away at the emotional peak. Keep every cutaway honest: it must belong to the world of the film.
Scene / line: ______ Weak moments (time ranges): ______ Shot # | Picture (speaker / reaction / cutaway / wide) | In - Out | Audio under it 1 | | | 2 | | | 3 | | | Sound bridge (line, shot to shot): ______ Supporting shots I still need to create: ______ What the audience will be told about synthetic performance: ______
Studio task: Plan and create the supporting shots needed to hide one limitation honestly.
Build the revised dialogue edit plan for your own shot.
Checklist before you submit:
Name the cutaway that best preserves meaning.
Deliverable: Revised dialogue edit plan.
Submit the revised dialogue edit plan. The file must use the course naming convention and include enough process evidence to show the student's decisions and revisions.
File name (course naming convention): AIF_S06-L04_revised-dialogue-edit-plan_v01_YOURINITIALS
| Criterion | Advanced | Proficient | Developing | Beginning |
|---|---|---|---|---|
| The work addresses the stated story or production purpose. | Every cut serves the story moment, and the plan shows why each supporting shot was chosen over alternatives. | The plan serves the story moment and the cuts have reasons. | The plan partly serves the story; some cuts lack reasons. | The plan does not connect to the story. |
| The student applies the required method accurately and produces usable capstone evidence. | Plan includes coverage, in and out times, a sound bridge, and at least one created supporting shot; another editor could follow it. | Plan includes coverage with times and one supporting shot. | Plan is incomplete or lacks times. | No usable edit plan. |
| The student identifies at least one quality issue and records a meaningful revision decision. | Tests the sequence at normal speed, names remaining flaws, and revises the cut points. | Identifies a quality issue and records a meaningful revision decision. | Notes a problem without revising. | No revision decision recorded. |
| The work complies with consent rights attribution privacy disclosure and school policy. | Disclosure plan stated; every supporting shot has documented provenance; no deceptive concealment. | Disclosure plan stated and provenance recorded. | Disclosure or provenance is missing for part of the plan. | Edit conceals required disclosure or uses unauthorized material. |
Requirement: Editing may solve quality problems but must not conceal required AI disclosure.
It is good craft to cover a weak frame. It is not acceptable to use editing to hide that a performance is synthetic when disclosure is required. Cut for quality, and label for honesty. Your final film credits or notes should say where AI-generated faces or voices were used. Any supporting shots must also be created by you, licensed, or released in writing by the people who appear in them.