Highlight Sweep C04
0.0s
Previews are illustrative demonstrations of the technique — not your footage. Your real edit is produced later by Claude Code.
C04 · Captions · intermediate · 5–180s
Full caption lines stay steady while an accent-coloured block sweeps in behind one key word per line exactly as it is spoken.
The caption reads like a normal subtitle, but selected words get a marker-style highlight that wipes in from left to right on their spoken frame, flipping the word to a dark ink for contrast. Limiting it to one or two words per line turns the captions into an emphasis system: the viewer's eye is pulled to the word that carries the meaning, much like a highlighter on a printed page.
Try the parameters
Distance from the platform UI edge, as % of frame height.
Best used for
- — Educational explainers
- — Business and finance tips
- — Key-term heavy scripts
- — Course clips
Assets
- ● Source video required
- ○ Transcript with timing optional
- ○ Brand fonts optional
Pairs well with
If assets are missing
If keywords can't be identified with confidence, ask the user to supply them. Without word timing, transcribe locally first; as a last resort apply highlights at the start of each line rather than mid-line on guessed timing.
Accessibility
Text on the highlight must hold 4.5:1 against the highlight colour (dark ink on light accents). Don't rely on highlight colour alone to convey meaning — the words remain readable without it.
Individual prompt
updates with the parameters above# CES style instruction — C04 Highlight Sweep You are an experienced motion designer and video editor working in Claude Code. Apply the **C04 Highlight Sweep** technique to the supplied footage as specified below. **B. Project configuration:** 1080×1920, 9:16, 30 fps · Source video: not provided — ask the user for the path before starting · Video language (on-screen text & captions): English **Captions** — The caption reads like a normal subtitle, but selected words get a marker-style highlight that wipes in from left to right on their spoken frame, flipping the word to a dark ink for contrast. Limiting it to one or two words per line turns the captions into an emphasis system: the viewer's eye is pulled to the word that carries the meaning, much like a highlighter on a printed page. ### C04 Highlight Sweep - **When:** Applies to the whole video - **Layer / track:** Captions layer (topmost text) - **What it does:** Full caption lines stay steady while an accent-coloured block sweeps in behind one key word per line exactly as it is spoken. - **Parameters:** Highlighted words: (not set); Highlight colour: #D7FF5F; Text on highlight: #0B0C0E; Highlight shape: Solid block; Sweep duration: 0.2 s; Max highlights per line: 1; Caption typeface: Bold sans (Space Grotesk / Inter Black); Safe-zone offset: 18% - **Steps:** 1. Add a highlight-sweep caption layer across every spoken section of the video on the top track (1080×1920 @ 30fps). 2. Timing source: use the supplied transcript's word-level timestamps for the English dialogue. If it only has line/sentence timing, or no transcript was supplied, transcribe the source audio locally with Whisper word timestamps (e.g. `whisper source.mp4 --word_timestamps True --output_format json`, or WhisperX for tighter forced alignment) and save the result as captions.json. Caption text is the speaker's words verbatim — never paraphrase, "correct", summarise or invent wording; list any low-confidence words for the user to confirm. 3. Choose highlight words from the transcript itself: the noun, number or verb that carries each sentence's meaning (as spoken). Never add words the speaker did not say. 4. Captions: lines of at most 26 characters, two lines per page, pages cut on phrase boundaries; text in Space Grotesk at Black/ExtraBold weight (fallback: Inter Black), cap height ≈ 6.5% of frame width, colour #F4F2EB with a soft shadow. Highlight at most 1 word per line. 5. Sweep: on the highlighted word's start frame, draw a solid rectangle 1.15× the word's cap height, padded 0.12em horizontally, radius 0.08em in #D7FF5F behind the word, scaling X from 0 → 1 from its left edge over 6 frames using a fast expo-out curve (cubic-bezier(0.16, 1, 0.3, 1)); switch the word's fill to #0B0C0E once the band passes its midpoint. Highlights persist until the page clears. 6. Placement: anchor the caption block's bottom edge 18% of frame height (346px at 1080×1920) above the bottom of the frame, horizontally centered with ≥ 8% side margins, so it stays clear of the TikTok / Reels / Shorts interface (description text and action buttons occupy roughly the bottom 20% and the right-hand 12% of the frame). Never cover the speaker's eyes or mouth; if the face sits unusually low in a shot, lift the block above it for that shot only. 7. Implement as a Remotion <HighlightSweep /> component: measure each word box with a hidden layout pass (or @remotion/layout-utils measureText) so the band matches the rendered glyph width exactly; render band layer under text layer inside the same line container. - **Assets:** Source video (⚠ MISSING) · optional: Transcript with timing, Brand fonts - **Fallback:** If keywords can't be identified with confidence, ask the user to supply them. Without word timing, transcribe locally first; as a last resort apply highlights at the start of each line rather than mid-line on guessed timing. **G. Asset integrity and restrictions:** - Do not create or present invented testimonials, comments, quotes, reviews or endorsements. - Do not fabricate numbers, metrics, charts or results. Data visualisations use only the supplied data. - Do not create fake screenshots or UI and present them as real. Illustrative graphics must be clearly illustrative. - Do not attempt unsupported 3D reconstructions or perfect subject separation without a mask; use the documented 2D fallbacks instead. **I. Quality verification:** - Check output dimensions and aspect ratio with ffprobe — they must match the configuration exactly. - Check the output frame rate and that the duration is within ±0.5 s of the target. - Check audio/video synchronisation at the start, middle and end (lip-sync and SFX hits within 1 frame). - Inspect frames at every beat boundary for text clipping, overflow and safe-zone violations. - Confirm the output file exists and plays — never claim a render is complete without an actual output file. - Report any missing asset or unimplemented effect explicitly.
More captions styles
Choose one primary caption style per video. It usually runs across all spoken sections as a global layer.
- C01Karaoke FollowShort caption pages sit over the speaker and each word fills with the accent colour at the exact moment it is spoken.
- C02Pop Word CaptionsOne to three words at a time pop onto the speaker's chest with a small scale overshoot and a short upward drift, then hand off to the next.
- C03Typewriter SubtitlesEach subtitle types itself on character by character with a blinking caret, paced so the phrase completes as the speaker finishes it.
- C05Stacked CaptionsEach phrase is set as a tight stack of two or three short lines that rise into view one after another, the key line set larger.
- C06Editorial Lower ThirdsCalm sentence-case subtitles sit low-left in a refined typeface under a thin accent rule, crossfading softly between sentences.
- C07Explosive KeywordsA regular caption line runs until a key word bursts out of it, filling the frame at several times its size for a beat before snapping back.
- C08Subtitle CapsulesEach short phrase appears inside a rounded label capsule that stretches, pops or slides to fit the next phrase as the speaker talks.
- C09Monospace Terminal CaptionsDialogue prints word by word in a monospace terminal block with a prompt and blinking cursor, older lines scrolling up and dimming.
- C10Bilingual Caption FlowA bold primary caption in the spoken language is paired with a smaller, differently coloured translation line, both kept inside the safe zone.