Pop Word Captions C02

Pairs well with

# CES style instruction — C02 Pop Word Captions

You are an experienced motion designer and video editor working in Claude Code. Apply the **C02 Pop Word Captions** 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** — Instead of a full sentence, the viewer sees only the word or two being spoken right now. Each group arrives with a restrained scale overshoot and a few percent of upward travel, outlined for contrast, and is replaced the moment the next group starts. The constant small motion keeps eyes anchored near the face and gives fast-paced delivery a punchy, rhythmic feel without becoming chaotic.

### C02 Pop Word Captions

- **When:** Applies to the whole video
- **Layer / track:** Captions layer (topmost text)
- **What it does:** One 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.
- **Parameters:** Words per pop: 1 word; Pop overshoot: 1.12x; Rise distance: 2%; Easing: Back out (overshoot); Caption typeface: Bold sans (Space Grotesk / Inter Black); Text size: 9%; Emphasis colour: #D7FF5F; Safe-zone offset: 22%
- **Steps:**
  1. Add a pop-word 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. Grouping: show exactly one word at a time (max 14 characters); merge articles and short function words ("a", "the", "to") with the following word so nothing flashes for under 4 frames. A group appears on its first word's start frame and is replaced when the next group starts; clear after 15 frames of silence.
  4. Entrance per group: scale 0.72 → 1.12 → 1.0 and translate up from +2% of frame height to rest over 9 frames using a back-out curve with slight overshoot (cubic-bezier(0.34, 1.56, 0.64, 1)); opacity 0 → 1 in the first 2 frames. No exit animation — the next group replaces it on a hard cut.
  5. Typeset in Space Grotesk at Black/ExtraBold weight (fallback: Inter Black), cap height ≈ 9% of frame width (≈ 97px), with a solid dark outline (≈ 0.14em, paint-order stroke) for contrast. Colour the single most important word of each sentence (the noun or number the sentence is about, as spoken) in #D7FF5F — at most one emphasis per sentence.
  6. Placement: anchor the word group's bottom edge 22% of frame height (422px 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 <PopWords /> component reading captions.json; compute each group's local frame (frame − startFrame) and drive scale/translate with interpolate() or spring(). Check that no group is visible for fewer than 4 frames.
- **Assets:** Source video (⚠ MISSING) · optional: Transcript with timing, Brand fonts
- **Fallback:** Without word timing, transcribe locally with Whisper before building this style. If that is impossible, use C06 Editorial Lower Thirds with line-level timing instead — word pops on guessed timing look visibly out of sync.

**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.