The Voice Director Workbench
A one-minute product launch film turns voice creation into a creator's sequence of audible revisions, UI choices, and a late trust payoff.
Field notes / why it works
The creator states a real production task in the first four seconds, then every new spoken instruction receives a visible UI consequence: three options by 0:13, a library at 0:21–0:25, and parameter controls from 0:28 onward. Only 10 certain cuts occur in 60 seconds, but 15 internal graphic steps keep the proof moving. The 0:50–0:51 consent-to-ready reveal is the strongest replay peak, so the final feature also functions as a payoff.
Plate 01 / The format in one breath
A creator at a laptop tries to make a piece of content, directing the product aloud as if it were a collaborator. Each request cuts to a clean interface or visual metaphor that makes the feature visible; the demonstration escalates from making a first result to customizing it, combining options, and protecting a personal input. The promise is practical control over a complex creative tool, expressed as a short story of revisions.
Why it works
- The hook gives a concrete job at 0:00 and names the first step by 0:03; a single locked presenter shot lets the viewer understand the premise before the first cut at 0:04.4.
- The product proof changes form as the job advances: typed direction at 0:10, three generated options at 0:13.6, a voice library at 0:21–0:25, and a waveform with stacked controls at 0:37–0:41.
- The final new capability arrives late: personal voice at 0:43, followed by a verification-to-ready transition at 0:50–0:51. The latter is the strongest replay peak, 0:51.0–0:51.6.
Plate 02 / Format card
| Platform · aspect · length | YouTube · 16:9 · 60 s (source measured at 640×360, 25 fps) |
| Pace | 11 shots · 10–11 cuts/min including one uncertain change · 25 picture changes/min with UI steps · mean shot 5.46 s, median 3.44 s |
| Script | 157 measured words · 163 wpm overall, 192 while speaking · 18 short sentences averaging 8.7 words · one question |
| Voice | Low-register, quick, animated adult presenter; conversational direction alternates with polished demonstration lines. Other voice samples are character archetypes, not impersonations. |
| On screen | About 35% direct presenter, 40% full-frame UI/graphics, 25% overhead live action with product cards (estimated from shot durations and the mixed presenter/UI shot). |
| Captions | No running burned subtitles visible; product labels and demo copy carry the information. |
| Sound | Quiet tonal bed roughly 16.5 dB below voice (music-like, inferred); no consistent cut SFX; −15.3 LUFS, −1.0 dBFS true peak. |
| Structure | Seven beats: task → first try → revision → options → deeper control → personal example and trust → brand line. |
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- Plate 03 / The hook, frame by frame (first 3 seconds)Pro
- Plate 04 / Structure (beat sheet)Pro
- Plate 05 / Shot grammarPro
- Plate 06 / Visual style guidePro
- Plate 07 / Voice and deliveryPro
- Plate 08 / Sound designPro
- Plate 09 / PackagingPro
- Plate 10 / Script templatePro
- Plate 11 / Production planPro
- Plate 12 / PromptsPro
- Plate 13 / Edit recipePro
- Plate 14 / QA checklistPro
- Plate 15 / Make it yoursPro
- Recipe / runnable promptPro