Showcases

From Text to Storyboard: A Multi-Shot Narrative Translation Example

Taking "how a passage of text is translated into imagery" as its premise, this demonstrates the text-to-video path of Seedance 2.0 fast: the shot breakdowns, progression of shot sizes, and subject-consistency requirements are written explicitly into a single text prompt, letting the model organize a multi-shot sequence from environmental establishing to detail close-up within a single generation of about 8 seconds, with native audio enabled to verify audio-visual sync. The piece deliberately introduces no reference material, and is intended to observe whether shot segmentation under pure text input follows the prompt's structure. It is suited to short-video narrative previsualization, advertising storyboard concept validation, and developer evaluation of prompt controllability in text-to-video.

Style Storyboard · Narrative · Cinematic

Model Seedance 2.0

Try it yourself

Prompt

Try this model

A short multi-shot cinematic sequence generated from a single text prompt. Shot 1 (wide, 2 seconds): a quiet, rain-soaked city street at blue hour, sodium lights reflecting in puddles, a haze of traffic in the distance. Shot 2 (medium, 2 seconds): a lone courier in a dark teal raincoat steps off the curb, breath visible in the cold air, the raincoat and bag consistent across shots. Shot 3 (close-up, 2 seconds): water droplets fall from the edge of the courier's hood and strike a reflective watch face, shallow depth of field. Shot 4 (insert, 2 seconds): the courier's hand opens a sealed envelope, warm interior light spilling across the paper. Camera language: a steady lateral dolly in Shot 1, a subtle handheld follow in Shot 2, a locked-off macro in Shot 3, and a gentle push-in in Shot 4. A cool blue-teal palette with warm amber accents, cinematic contrast, natural motion blur, a 24fps filmic texture, and a consistent subject design across all shots.

Copy the full prompt and adapt it for your own generation.