The project
Use a short interview or camera-to-viewer recording, two or three relevant B-roll shots, and optional licensed music. The goal is a clear one-minute sequence, not the maximum number of cuts. Complete the quick start first if you are new to the timeline.
1. Build a speech-only cut
Import the recording, inspect its sound, and create a sequence. Listen once without editing and identify the main promise, supporting points, and final takeaway. Mark the useful source ranges and append them in a logical order.
Keep enough context that the speaker’s meaning remains accurate. Removing a hesitation can improve clarity; removing a qualification can change the meaning. Rewatch the original around any ambiguous cut.
2. Tighten with intention
Select a clip, position the playhead, and split with ⌘K. Ripple delete unwanted material when the gap should close. Leave natural breaths and pauses where they help the audience understand a new idea.
Text-based editing can help locate silences or filler words. Treat its suggestions as candidates and listen to each result; the correct rhythm depends on the speaker and sentence.
3. Cover the edit with useful picture
Place a relevant B-roll range above the spoken sequence. Keep the dialogue continuous underneath. Enter the new image near the word it explains, and return to the speaker when their expression adds information.
4. Finish the sound and captions
Balance dialogue clips before adding music. Use a modest music level and review ducking around quiet phrases. Generate or import captions, correct the transcript, and inspect their timing.
Review your result
Watch once for meaning, once for rhythm, and once for sound and captions. Check every removed phrase against the original when necessary. Save a named version of the approved cut before making a vertical adaptation.