Skip to content
refs
REF-2986Source onlineBlueprint rev 01

Scribe v2 Realtime launch film

A restrained speech-to-text launch film that turns accuracy claims into comparison graphics and a live transcription scenario.

Maker @ElevenLabsFor ElevenLabsOriginal post ↗
Cut timeline · 4 shots01:21
00:0000:1500:3000:4501:0001:15

Field notes / why it works

The staged blank-to-mark-to-title opening resolves the product name by 2.75 seconds, while the spoken claim leads into a comparison graphic by about 7 seconds. An everyday message exchange at 32–40 seconds and streaming UI details at 42–60 seconds turn an invisible technical benefit into a visible workflow. Eleven internal graphic changes sustain interest despite only three detected hard cuts in 81.5 seconds.

01

Plate 01 / The format in one breath

A calm, almost white launch film announces a technical product, asserts one benefit, shows evidence, then makes the benefit tangible through a miniature interface scenario. Sparse typography and precise motion give the spoken claims room to register; the final line directs viewers to build with the product.

Why it works

  • The first 2.5 seconds move from blank field to mark to full product name, while the narrator names the product by 1.5 seconds. The staged reveal makes a conventional announcement feel deliberate.
  • An abstract accuracy claim at 3–7 seconds receives a proof visual at 7–14 seconds, then a concrete chat example at 32–40 seconds. The film changes the kind of evidence before the claim grows stale.
  • At 42–60 seconds, streaming text, a recording control, chat, waveform, and timing ruler make an invisible speech technology visible. The film can stay visually quiet while still adding new information.
02

Plate 02 / Format card

Platform · aspect · lengthX · 1280 × 720, 16:9 · 81.5 s · 60 fps
PaceDetection groups the film into 4 shots, 3 hard cuts (2.2/min), plus 11 internal changes; about 10.3 total picture changes/min. Mean detected shot 20.38 s; most motion happens within scenes.
Script177 spoken words · 138 wpm overall, 159 while talking · 13 sentences, 13.6 words each · 2 questions · 10 number mentions
VoiceLow-register, composed product narrator; clearly enunciated, with animated pitch on the claims
On screenDesigned typography, abstract motion graphics, benchmark chart, simulated messaging and recording UI; no visible presenter or live action
CaptionsNo continuous subtitles; selective centered headline and interface text in black or charcoal
SoundQuiet tonal music-like bed (automated classification; inferred ambient electronic mood), about 3.5 dB below voice by the measurement; no identifiable cut effects in supplied facts
StructureReveal → claim and comparison → mechanism → use case → feature and trust cards → CTA and logo

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 frame (first 3 seconds)Pro
  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