AI Video Prompt Translator

Describe a shot once in plain words. Get a ready-to-paste prompt for Kling, Seedance, MiniMax, LTX, Wan, FLUX and Gemini Omni, each written the way that model was trained to read it.

Free7 models at onceDialogue written in each model's own format
Who is in it, what happens, where, and how the camera moves. Plain words are fine.
Add a spoken line (optional)
6 s
Aspect ratio
Models

Real Example

One Shot, Seven Dialects

This is a real result from the translator, unedited. Notice how the same shot becomes structured fields for MiniMax, bracketed sound channels for Seedance, one flowing paragraph for LTX and a silent version for Kling.

The shot

Night, rain-streaked diner window. A woman in a red coat sits alone in the last booth, turning a ring on the table. Slow push in from the window.

Dialogue: Mara: "You kept it."

6 s, 16:9

Kling 3.0

SilentThis 6 s shot: $1.12 at 1080p (billed as 10 s)
Woman in a red coat, dark hair, sits alone in the last booth of a diner at night, rain streaking down the window beside her. She slowly turns a ring between her fingers on the table, eyes lowered, lips pressed into a quiet, guarded line. Camera performs a slow dolly in from outside the window, moving through the rain-streaked glass toward her face and hands. Warm interior light glows against the cold blue night outside. Mood: tender, melancholic stillness.

Line left out.Kling 3.0 renders without speech, so the spoken words are not in this prompt. Add the voice in the edit, or pick a model that speaks.

  • Pick the 5s option and let it play near-full length since the push is gentle; avoid padding with extra movement.
  • Watch that rain refraction on the glass doesn't smear her face too much as the lens gets close - keep the push modest.

Seedance 2.5

Speaks dialogueThis 6 s shot: $2.84 at 720p
A woman in a red coat sits alone in the last booth of a diner at night, slowly turning a ring between her fingers on the table. Rain streaks down the window beside her, blurring the glow of the street outside; warm amber light from the diner interior contrasts with the cold blue night beyond. Slow push in from outside the window, moving through the rain-streaked glass until it settles close on her face and hands. Dialogue language: American English. Mara says quietly: {You kept it.} <soft rain ticking on glass> <faint diner hum> (sparse, melancholic piano underscore, soft dynamics)
  • Keep this as one continuous 6-second take - no need for staged beats at this length.
  • Time the line so her lips move right as the push settles close, so the rain-glass blur doesn't fight the lip-sync.

MiniMax H3

Speaks dialogueThis 6 s shot: $0.36 at 768p
For the target video, at 0.00 seconds into the target video, <Picture 1> (from [Shot 1]) is fully referenced.

integrated_multimodal_description: [Shot 1] Live-action, cinematic night diner scene. A woman in a red coat sits alone in the last booth, rain streaking down the window beside her, warm interior light glowing against the cold dark outside. She turns a ring slowly between her fingers on the tabletop, eyes lowered. The camera pushes in with small amplitude at slow speed, moving from outside the rain-streaked window toward her face and hands. As it settles close, the young woman with a quiet, steady voice (S1) says: <d>[English] You kept it.</d>

overall_soundscape: Rain ticking softly against the glass, a faint hum of refrigeration or a neon sign, the low murmur of an otherwise empty diner, and a soft clink as the ring taps the tabletop.

non_diegetic_music: A sparse, melancholic piano motif, slow tempo, soft dynamics, underscoring the quiet moment.
  • Keep the push amplitude small since the clip is only ~6s - a big move will overshoot before she speaks.
  • Reference the booth and window clearly as Picture 1 so the push-in direction matches the existing frame.

LTX 2.5

Speaks dialogueThis 6 s shot: $0.72 at 720p
A slow push-in through a rain-streaked diner window at night reveals a woman in a red coat sitting alone in the last booth, warm interior light glowing against the cold blue dark outside. Water runs in thin rivulets down the glass, softly distorting her silhouette as the camera gradually moves closer. She sits with her shoulders curled inward, slowly turning a ring between her thumb and forefinger on the tabletop, her gaze fixed downward, jaw tight. Her dark hair falls loosely around her face, the collar of her red coat damp at the edges. As the camera settles closer, her expression comes into focus, quiet and unmoving, before she lifts her eyes. She says in a soft, flat American accent, "You kept it." The hush of rain tapping the window blends with the low murmur of an empty diner and the faint hum of a distant neon sign.
  • Keep the push subtle across the 6-8s runtime so the glass and her face don't blur together too fast.
  • Hold one consistent light logic - warm interior vs cool exterior - so the push-in doesn't flatten the contrast.

Wan 3.0

Speaks dialogueThis 6 s shot: $0.60 at 720p
A woman in a red coat sits alone in the last booth of a rain-streaked diner at night, slowly turning a ring between her fingers on the table, her eyes lowered in quiet thought. Outside the window, rain runs down the glass in thin streaks, blurring the soft glow of a distant neon sign; warm diner light spills over her face and hands. The camera performs a slow push in from outside the window, gliding through the rain-streaked glass until it settles close on her hands and face. As it nears, she lifts her eyes and says, quietly, "You kept it." Rain ticks steadily against the glass, the diner's ambient hum and a distant clink of cups fill the quiet background.
  • At 6 seconds, keep the push slow and short so it doesn't rush past her face before she speaks.
  • Let the rain drops on the glass read as soft foreground blur early in the move, clearing slightly as the camera nears her.

FLUX 3

Speaks dialogueThis 6 s shot: $1.02 at 720p
Slow push-in shot of a woman in a red coat sitting alone in the last booth of a rain-streaked diner at night, slowly turning a ring between her fingers on the table. Warm interior light glows against the cold, rain-streaked window, her eyes lowered in quiet thought as the camera moves closer from outside the glass. Rain ticks softly against the window, and the low hum of an empty diner fills the background. She lifts her eyes and says, "You kept it." (no on-screen text, no subtitles)
  • Use just the one movement term, slow push-in, so the model doesn't blend in extra camera motion.
  • Repeat this exact subject description word-for-word if you generate further shots of her for continuity.

Gemini Omni

Speaks dialogueThis 6 s shot: $0.60 at 720p
A slow push-in shot moves through a rain-streaked diner window toward a woman in a red coat sitting alone in the last booth at night. She slowly turns a ring between her fingers on the table, eyes lowered, as warm amber light from the diner interior contrasts with the cold, rain-blurred night outside. As the camera nears, she lifts her eyes and says quietly: "You kept it." Soft rain ticks against the glass, a low diner hum and distant clink of cups fill the quiet space, with a faint, melancholic ambient tone underneath.
  • Keep the clip around 6-8 seconds so the single quiet line has time to land before the shot ends.
  • 16:9 suits the window framing - make sure the push-in stays centered on her so the booth edges don't crop oddly.
The Method

Why One Prompt Does Not Fit Every Video Model

Every video model was trained on captions written in a particular way, and it follows prompts best when they look like those captions. A prompt that works beautifully on one model can be ignored, misread or even spoken aloud by another.

  • MiniMax H3 expects a strict structure: an instruction line, then labelled fields for the picture, the soundscape and the music, with dialogue in tags.
  • Seedance 2.5 reads sound from bracket channels: music in round brackets, sound effects in angle brackets and dialogue in curly braces.
  • LTX 2.5 wants one flowing paragraph of present-tense prose with no tags at all, and emotion shown as physical cues.
  • FLUX 3 reads like a director's note: camera, subject, action, place, then the sounds tied to what is on screen.
  • Kling 3.0 follows its own formula of subject, movement and scene, then camera, light and atmosphere, and renders without speech.
  • Wan 3.0 and Gemini Omni take natural prose with spoken words in quotes; Gemini always makes sound, so a silent shot has to say so.

Getting the Most From Your Prompts

  • Start from a good frame. These are image-to-video prompts. The picture already shows what things look like, so the prompt spends its words on movement, camera and sound.
  • Match the clip length. Set the length you will generate. Some models sell fixed lengths; the cost line on each card shows when a shot is rounded up.
  • Keep the exact words. Spoken lines are passed through word for word, in each model's own format. Silent models leave the line out rather than mouthing words with no sound.
  • Change one thing at a time. If a take misses, adjust the camera move or the action, not everything at once.
Next in PyxelJam Studio

Stop rewriting the same shot for every model.

In PyxelJam Studio every shot gets its prompt written for whichever video model you pick, with dialogue, camera and lighting in that model's own wording, straight from your script.

Make this shot in PXLJ Your shot description comes with you.

Illustration of what comes next. Real frames are drawn from your own shots after you sign up.

ScriptShot listFramesClipsFinal cutPublish

Questions

What does an AI video prompt translator do?

It takes one plain description of a shot and rewrites it in the prompt format each video model was trained to read: structured fields for some, bracketed sound channels for others, one prose paragraph for others. The shot stays the same; only the wording changes.

Which video models does it write prompts for?

Kling 3.0, Seedance 2.5, MiniMax H3, LTX 2.5, Wan 3.0, FLUX 3 and Gemini Omni. The model spec sheet lists what each one can do.

Does it handle dialogue and lip-sync?

Yes, for the models that speak: Seedance 2.5, MiniMax H3, LTX 2.5, Wan 3.0, FLUX 3 and Gemini Omni. Each gets the exact words in its own dialogue format. Kling 3.0 renders without speech, so the line is left out of that prompt rather than having a character mouth words with no sound.

Do I need a start frame?

The prompts are written for image-to-video: you give the model a picture and the prompt describes what happens next. That is why they focus on movement, camera and sound rather than describing how everything looks.

Can I use these prompts in any app?

Yes. Paste them into any tool that runs these models. They are plain text and free to use.

Is it free?

Yes, with a daily allowance per visitor so it stays free for everyone. Leave your email on a result and the allowance goes up.