Skip to content

SCRIPT.md

Latest commit

 

History

History
92 lines (61 loc) · 5.31 KB

SCRIPT.md

File metadata and controls

92 lines (61 loc) · 5.31 KB

HyperFrames Launch Video — Script v3

Status: Approved copy. Awaiting VO generation (Phase 2). Relationship to v2: v2 is the recorded script in assets/voiceover.mp3. v3 supersedes it — contrast beat cut, cold open rewritten, flex section repositioned, anatomy copy updated (GSAP → JavaScript), engine copy revised, brand restyled to HyperFrames (camelCase).

Spoken VO

Spoken lines only. No inline directions, no emphasis markdown, no author notes. This block is what feeds ElevenLabs verbatim.

Imagine you could make videos like this.

Your agent already can — just give it HyperFrames.

Open source. HTML in, video out.

CSS animations. G-Sap. Lottie. Shaders. Three.js.

A div is a keyframe. Data attributes are your timeline.
CSS is your look. JavaScript is your animation engine.

Anything a browser can render can be a frame in your video.

Drop in music, sound effects, footage — it all composes together.

Your agent writes code to compose. HyperFrames renders every frame to MP4.
It's deterministic, pixel-perfect, every time.

Give your agent the skill. Tell it what to make. Watch it build.

HyperFrames. Go make something.

Silent visual beats (not spoken)

The VO leaves room for two silent gaps where the picture carries the moment. Not part of the ElevenLabs input — handled by Phase 3 timeline sequencing.

  1. After "Imagine you could make videos like this." — ~3s silent beat. Super-tight motion-graphics montage (existing flex reel content smashed together as a teaser before the product is named).
  2. After "Open source. HTML in, video out." — A-roll beat with composition overlay (creator on camera + lower third / captions / waveform). This is the Seedance insert from the original storyboard (Beat 6F in v2, Priority 3 in HANDOFF.md), now promoted earlier in the structure.

Pronunciation cues for VO generation (Phase 2)

For ElevenLabs. Do not paste these cues into the script block above — they are a separate instruction set.

  • HyperFrames — pronounced hyper-frames, not hyper then frames spelled out. The camelCase is a typographic choice, not a pronunciation cue. Expect ElevenLabs to handle it as a single compound word.
  • GSAP — pronounced G-Sap (rhymes with snap). Already inlined as "G-Sap" in the spoken block above for TTS.
  • Three.js — pronounced three dot J-S.
  • MP4 — pronounced M-P-four.

Changes vs. Jake's draft (the copy dialed in this pass)

From (draft) To (v3) Reason
"Imagine you could make videos like this" "Imagine you could make videos like this." Terminal period before the montage cut.
"Your Agent already can, just give it HyperFrames" "Your agent already can — just give it HyperFrames." Em-dash (per Jake). Lowercase "agent" for consistency.
"CSS animations (update). GSAP. Lottie. Shaders. Three.js." "CSS animations. GSAP. Lottie. Shaders. Three.js." "(update)" was an author note. Dropped.
"Drop in music, sound effects, footage — it all composes together." "Drop in music, sound effects, footage — it all composes together." Asterisks were markdown emphasis. Stripped. Emphasis goes into delivery.
"renders every frame into MP4" "renders every frame to MP4" to is more idiomatic than into for single-format output.
"pixel perfect" "pixel-perfect" Compound adjective, hyphenated.

Changes vs. v2 (the recorded VO)

v2 beat v2 VO v3 status
1 canvas-open "Your AI agent already knows how to make videos." Replaced — new cold open: "Imagine you could make videos like this."
2 canvas-zoom "It just needs the right format. This is Hyperframes." Folded into "Your agent already can — just give it HyperFrames."
3 proposition "An open source framework. HTML in, video out." Simplified — "Open source. HTML in, video out."
(new) — A-roll + composition overlay beat promoted from old Beat 6F.
6A–6E flex "CSS animations. GSAP. Lottie. Shaders. Three.js." Repositioned — now plays before anatomy. Also feeds the opening montage.
4 anatomy "…CSS is your look. GSAP is your animation engine." Copy change — GSAP → JavaScript.
5 thesis "Anything a browser can render can be a frame in your video." Preserved.
6F–6L compose "Drop in music, sound effects, footage — it all composes together." Preserved.
7 contrast "No new framework for the agent to learn…" CUT ENTIRELY. ~6.4s freed.
8 engine "The agent writes it. The renderer captures every frame as MP4. It's deterministic. Identical outputs, every time." Revised — "Your agent writes code to compose. HyperFrames renders every frame to MP4. It's deterministic, pixel-perfect, every time."
9 cta "Give your agent the CLI. Tell it what to make. Watch it build." Reverted — CLI → skill.
10 canvas-close "Hyperframes. Go make something." Preserved with brand restyle — "HyperFrames. Go make something."

Word count

Spoken v3: ~110 words. v2 was ~140 words. Expected spoken duration ~50–55s at Apple-keynote cadence. With the ~3s motion-montage gap and the A-roll gap added back in, total runtime lands roughly in the 58–62s range — same neighborhood as v2, but with a tighter cold open and no contrast beat.