Skip to content
refs
REF-2773Source onlineBlueprint rev 01

Just Speak: A Minimal Codex Concept Film

A 21-second voice-input concept film that turns sparse kinetic type into a mic demo, then waits for a scenic Codex reveal.

Maker @shapelayerFor OpenAI CodexOriginal post ↗
Cut timeline · 5 shots00:21
00:0000:0500:1000:1500:20

Field notes / why it works

The promise completes on a nearly empty canvas by 0:00.75, then a large cursor and microphone make it tangible by 0:02. Only six spoken words are needed; the 0:05.2 interface glimpse and 0:08.8 white reset defer the branded payoff until 0:16.8.

01

Plate 01 / The format in one breath

A bare white canvas turns a short second-person instruction into a microphone control. One spoken request triggers a fleeting interface demonstration; the film then clears to a tiny mark before resolving into a scenic brand card. The withheld product name makes the last five seconds the answer to the opening question.

Why it works

  • The first frame contains only one blue word; the complete promise arrives by 0:00.75, without requiring the viewer to parse a busy UI.
  • An oversized cursor and blue microphone pill turn the 0:01.5–0:05.2 instruction into a physical action. The only speech, six words at 0:01.66–0:02.90, demonstrates the same action.
  • The visual scale pivots twice: an interface glimpse at 0:05.2–0:08.8 and a minimal white reset from 0:08.8–0:16.8. This reserves the flower-sky brand payoff for 0:16.8–0:21.4.
02

Plate 02 / Format card

Reference
Platform · aspect · lengthX · 1280×720, 16:9 · 21.4 s at 29.97 fps
Pace5 measured shots; 4 cuts; 11.2–14.0 possible cuts/min; 16.8 measured picture changes/min; mean shot 4.28 s. Text morphs inside shots add motion below the change detector's threshold.
ScriptOne six-word spoken question, first word at 0:01.66; 243 wpm during that brief utterance; speech covers 7% of runtime.
VoiceOne medium-register, animated, conversational request; no narrator after 0:03.
On screenFull-frame kinetic type and microphone UI → soft-focus interface footage → white symbol morph → scenic product end card. No person appears.
CaptionsTiny lower-centre transcription of the request during the mic shot; the large opening words are headline typography, not speech captions.
SoundMusic-like bed (inferred from tempo estimate, about 87 BPM); one spoken line; no clear cut-synced effects. −10.6 LUFS integrated, +0.3 dBFS measured true peak.
Structure5 beats: instruction, spoken request, interface, abstract reset, product reveal. Payoff begins 0:16.8; no explicit CTA.

Pro / 13 plates + recipe

The first plates are open. The full dissection is Pro.

Get the timed shots, type and sound specs, and the runnable recipe.

Open the full blueprint
  1. Plate 03 / The hook, frame by framePro
  2. Plate 04 / Structure (beat sheet)Pro
  3. Plate 05 / Shot grammarPro
  4. Plate 06 / Visual style guidePro
  5. Plate 07 / Voice and deliveryPro
  6. Plate 08 / Sound designPro
  7. Plate 09 / PackagingPro
  8. Plate 10 / Script templatePro
  9. Plate 11 / Production planPro
  10. Plate 12 / PromptsPro
  11. Plate 13 / Edit recipePro
  12. Plate 14 / QA checklistPro
  13. Plate 15 / Make it yoursPro
  14. Recipe / runnable promptPro