Seedance Video Prompt Template With Timecoded Beats
A detailed video prompt-writing template covering recurring character blocks, reference asset bindings for images, video and audio, timecoded beats, transition directives, consistency locks, default exclusions, plus guidance on dialogue, layered audio, extensions and edits, with two full worked examples.
Source previews
These example images come from the original source. DROPIDEA has not independently tested these examples; results depend on the model and settings.
Your prompt
Replace the red variables with your information. Copy the full prompt and use it with your preferred agent.
Variable guide
[CHARACTER]Fixed character-block tag; write the physical description after it
Line 'Optional [CHARACTER] block' and the second example
Example value[CHARACTER] Man in his forties, short grey hair
[SOUND]Fixed literal tag for excluding music; keep the whole quoted line unchanged
'Forbidden items' section and end of example 1
Example value[SOUND] Strictly only naturally occurring sound and foley, no music allowed.
[MUSIC]Music-block tag; fixed, describe one continuous composition after it
'Forbidden items' section and end of example 2
Example value[MUSIC] light lo-fi beat, soft keys
[SCRIPT]Script-block tag placed before the first timecoded beat
Example 2
Example value[SCRIPT] 0:00–0:06 | ...
<Role>Character's role in the image reference binding
Quoted line '@Image 1 defines <Role>'s face...'
Example value<Role> → the presenter
[hair]Hair locked from the reference image
Quoted line '@Image 1 defines...'
Example value[hair] → short black hair
[clothing]Clothing locked from the reference image
Quoted line '@Image 1 defines...'
Example value[clothing] → navy suit
[motion]Only the motion locked from the reference video
Quoted line '@Video 1 defines only the [motion].'
Example value[motion] → handheld walk-cycle
[music role]Role of the reference audio asset
Quoted line '@Audio 1 is the [music role].'
Example value[music role] → ambient lounge score
[X-Ys]Beat time window in seconds with western digits
Beat-format line and examples' beats
Example value[X-Ys] → [0-8s]
{dialogue}Spoken line in braces, language and accent stated before it
Beat-format line and dialogue guidelines
Example value{dialogue} → {Franchement, essayez-la.}
[the exact landing frame]Exact final frame of the beat
Quoted 'End state: [the exact landing frame].'
Example value[the exact landing frame] → plate alone under the heat lamp
[WHIP PAN / SNAP ZOOM / SMASH CUT / HARD CUT on (motion)]Transition style; choose one of the four with the motion description
Quoted transition-directives line
Example value[WHIP PAN / SNAP ZOOM / SMASH CUT / HARD CUT on (motion)] → SNAP ZOOM
[language]Forced language of on-screen text
Quoted 'All on-screen text must be in [language].'
Example value[language] → Arabic
[others]Languages excluded from on-screen text
Quoted 'No [others] or garbled characters.'
Example value[others] → English, French
What you get
- A complete timecoded video prompt template with explicit end states inherited by each following beat.
- A consistency-lock mechanism for identity, wardrobe and set layout across long and short runs.
- Layered per-beat audio description: voice quality foreground, medium texture, environment underneath.
- Two fully worked examples: a music-free kitchen scene and a French UGC ad with a [MUSIC] block.
Use cases
- Writing short-video prompts for generators like Seedance with precise timed rhythm.
- UGC ads with recurring presenters and lip-synced localized dialogue.
- Multi-beat narrative scenes requiring character, set and natural-sound consistency.
- Video extensions and targeted edits while preserving the base footage.
How to use
- Identify recurring characters and write a [CHARACTER] block with a precise physical description, locking identity, face, hair and wardrobe across all registers.
- For each reference asset provided, add one binding line (image, video or audio) within the limits of 1–8 subjects from images, 1–5 from video, 5–10s clips, one image per angle.
- Break the video into [X-Ys] beats with one main action per window, performance direction plus {dialogue} if any, layered audio, and an explicit end state per beat.
- Add transition directives when register or location changes, a closing consistency-lock line, and restate it word for word at top and bottom for runs over 30s.
- End with the exclusions that fit the brief: the [SOUND] line when music is unwanted, or a [MUSIC] block when it is, plus the default exclusions quoted unchanged.
Related prompts
30-Second Script: The Forest Checkpoint
A fully timestamped 12-segment comedy video script following a tiny backpacker stopped at a mushroom-guarded checkp...
Comparing GPT Image 2.5 Sunburst and Flare
A request to generate the same image with Flare and Sunburst and compare results using a complex prompt, with a sam...
Organized Shot Planning Before AI Prompt Generation
A preparation methodology covering angles, camera movement, lighting, lens, and actions per part, then handing the...
PHASEFOLD APERTURE Image Prompt Rewriter
A system template that rewrites the user's text, image, or both into one structured prompt for GPT Image Gen V2 or...