Smalltown Boy AI dance prompt for Kling: paste-ready text, no song names
Updated:
Short answer
Paste the prompt below into Kling, add your two photos as reference images and generate. It describes three shots and asks for a retro synth-pop instrumental with comic screams, and it names no song or artist, because tools filter those. Expect a few attempts, or let Tip Toe AI handle it.

The prompt
Image-to-video, 9:16, roughly 15 seconds. Filmed with an ordinary wide-angle phone camera; the picture fills the whole frame edge to edge, with no fisheye circle and no black border. Reference image 1 is the LEAD and reference image 2 is the PARTNER. Keep each one's face, fur or hair, colours, body shape and clothing exactly as in its own image. Never swap their looks, never merge them and never add a third character. If a subject is an animal, keep its species: it stands upright on its hind legs, presses both front paws to the sides of its head and opens its mouth wide in a long, dramatic yowl, like a person screaming in despair. Scene: a sunny suburban backyard in front of a bright red painted wooden fence, green lawn, blue sky. A static wide-angle phone camera at chest height, very close to the screamer, so the screamer's head and shoulders fill the foreground. No readable text or logos. Shot 1 (0-7 s): the LEAD sits very close to the lens in the foreground, clutching their head with both hands and screaming without a pause, mouth wide open, rocking in despair; around 4 s they stretch both hands toward the lens, then clutch their head again. Small in the background, the PARTNER dances energetically without stopping, full body in view. Shot 2 (7-14 s, hard cut): they swap. Now the PARTNER is very close to the lens, clutching their head and screaming without a pause, while the LEAD dances energetically, small in the background. Shot 3 (14-15 s, hard cut): a wide shot with both standing side by side, dancing together. Audio: an energetic, driving retro 1980s synth-pop instrumental with a punchy electronic beat, under the characters' own long comic screams (a cat yowls, a dog howls, a person wails "aaaah"). No lyrics, no words. Photorealistic, no text, no subtitles, no logos, no watermark.
Tips for using it
- Add the Lead photo as reference image 1 and the Partner photo as reference image 2. If your tool accepts only one image, use a photo that shows both and rename the roles in the first lines.
- Keep song and artist names out. Tools filter them and copied music can cause rights problems, so the prompt asks for an original-sounding instrumental. Add the trending sound in the app when you post.
- Tools change often. Check the clip length and aspect ratio your plan allows and edit the first line to match.
Questions people ask
Why is the song not named in the prompt?
Video tools filter or refuse song and artist names, so a named prompt may be rejected. This page is independent of the artists and of the people who made the original dance videos.
Will the result look exactly like the trend?
Probably not. Faces, motion and timing depend on the tool, the model version and your photos, so use the prompt as a starting point.
Do I need to write a prompt on Tip Toe AI?
No. Tip Toe AI runs a fixed scene: you add two photos and press the make button.
Make your own Smalltown Boy video
Upload two photos and make your own video. No subscription, no watermark.
Make my video