From One Historic Cry to a Battlefield Chorus
A sound-led history explainer turns one archival-sounding yell into a massed charge through spectrogram proof and layered audio.
Starter prompt / by Refs
The creator shared no short prompt. This is ours for making a film like it with Claude.
Recreate how {HISTORICAL_SOUND} really sounded in a 40-second audio history explainer, starting from the archival recording at {AUDIO_FILE}. Show a glowing magma-toned spectrogram on black with the pitch contour traced in white and elegant serif labels marking each phase of the sound. In Python, clean the recording, shift the voice to sound younger, then layer many varied copies with offsets in timing, pitch and distance to grow it from one voice to a crowd of thousands, updating the spectrogram and caption at each step. Render 1920×1080 with ffmpeg.Field notes / why it works
The first 2.5 seconds pair an isolated cry with a bright pitch-traced spectrogram, making the sound both unusual and legible. Method cards and before/after plots lead to a close individual at 0:18.2, then the switch to a wide charge at 0:27.7 makes the layered-audio scale-up feel earned. Ten shots and 12 internal steps give the 41.4-second piece a measured 30.4 picture changes per minute without crowding the audio.
Plate 01 / The format in one breath
Start with one surprising, isolated sound and a scientific looking visual that tells the viewer exactly what they are hearing. Show three concise processing steps, then turn that single source into a spatial crowd in a dramatized scene. The narrative escalates through sound scale rather than narration.
Why it works
- At 0:00 the headline identifies an unusual archival sound while a bright pitch trace lets viewers watch it unfold; the first visual change waits until 0:02.5.
- The 0:05.4–0:18.2 middle provides waveform and spectrogram proof plus brief method cards, so the later spectacle has a clear cause.
- A close human figure at 0:18.2 becomes a wide massed scene at 0:27.7; the shift in visual scale matches the growing number of sound layers.
Plate 02 / Format card
| Platform · aspect · length | X video · 16:9 · 1280×720 · 24 fps · 41.4 s |
| Pace | 10 shots; 9 measured cuts, 13.0–14.5 cuts/min including one uncertain change; 30.4 total picture changes/min; 4.14 s average shot |
| Script | No reliable spoken words. The detector transcribed a sustained yell as Japanese characters; 480 wpm is invalid. Text cards carry the explanation. |
| Voice | One rough, high, sustained archival-sounding human cry, then processed and multiplied cries; no narrator |
| On screen | About 29% spectrogram/waveform, 40% cinematic human/battle imagery, 30% black typography cards |
| Captions | No subtitles. White editorial serif cards, fine technical labels, sparse mono parameter tags. |
| Sound | The cry is the lead instrument. No music bed can be confidently identified from the supplied evidence. −14.0 LUFS integrated; −4.0 dBFS true peak. |
| Structure | Source → challenge → source layers → processing → one-person audition → scale-up → mass payoff → final thesis |
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 (0:00–0:03)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