UNLIMITED MUSIC VIDEO Breakdown (Prompts Included)
Ready to create a cinematic Y2K music video with perfect lip-sync? Follow the quick steps below and copy the exact prompts to recreate this viral music video in seconds.
Step 1:
Open pollo.ai, select Video Generation, and upload the reference photo(s) shown below.
After uploading the reference photo(s), select Video Generation. Choose Reference to Video, Here you can choose Seedance 2.5, Seedance 2.0, Minimax H3, or Wan 3.0 — any of them works. Next, paste the prompt below, and click Generate.
Reference photos
Screenshot example
Create Music Video on Pollo AI
Upload your character image and prompt to generate a polished, cinematic music short.
Try Pollo AI for Free
Image Prompt (Character):
Professional 4‑state character reference sheet, same 23‑year‑old Western European female rap‑pop performer, early‑2000s Y2K aesthetic, K‑pop level polish.
Layout: 16:9 wide canvas, left‑to‑right: ①head‑shoulder beauty close‑up(30% width) ②front full‑body ③90° left‑side full‑body ④back full‑body. 3 full‑body panels equal width, same character height, shared floor line, orthographic fashion turnaround, minimal perspective distortion.
Character: 170cm, slender light adult female build. Western‑European facial anatomy, blue‑grey deep‑set eyes, fair cool‑pink skin, subtle nose‑cheek freckles, realistic 22‑24yo skin texture, NO celebrity likeness.
Hair: short messy platinum‑blonde wolf‑bob above shoulders, dark roots, subtle faded dusty‑pink thin strands at temples, identical hairstyle across all 4 panels.
Glasses(mandatory all views): translucent smoke‑grey wraparound Y2K fashion glasses, 70% grey‑lilac tint lenses, eyes clearly visible, NOT sunglasses.
Makeup: Y2K soft grey‑lilac eyeshadow, fine lifted black eyeliner, dusty‑pink blush, translucent rosy‑nude glossy lips. Calm confident slightly‑challenging expression.
Outfit:
Top: fitted sleeveless washed‑ivory Y2K baby tee, high scoop neck, subtle diagonal seam, asymmetric short hem, small faded lavender‑pink abstract airbrushed chest graphic(NO text/logo). Back of top mostly plain ivory with tiny faded lavender mark upper‑back.
Skirt: low‑rise faded light‑wash asymmetric cargo denim mini‑skirt, small utility flap, thin silver hanging chain, silver rivets & buckle. Ends upper‑mid thigh.
Boots: matching solid smoke‑lilac knee‑high stretch‑leather boots, almond toe, 7cm kitten heel, back vertical seam, NO patterns/logos. NO socks/tights.
Accessories: small silver hoops, silver bracelet, dusty‑pink resin bangle, delicate silver rings & ear cuffs. NO bags.
Background: neutral medium‑grey seamless studio background with faint cool‑lavender gradient.
Lighting: early‑2000s digital‑camera flash + soft frontal fashion light, realistic material textures.
Negative: NO extra figures, 5th panel, 3‑quarter views, inconsistent face/hair/glasses/makeup/outfit, oversized build, long hair, opaque sunglasses, goth/cyberpunk clothes, readable text, logos, watermarks, anime style.
Image Prompt (Scene):
Empty cinematic outdoor location, NO people/human objects. Large completely drained swimming pool inside Southern California open landscape, luminous pink sunset sky. 16:9 wide.
Camera: elevated frontal‑oblique, 4‑5m height, 30‑35° downward angle, 28mm cinematic lens. Show full pool + large open sky, NOT top‑down / pool‑level / inside‑pool shot.
Sky & horizon: upper 35‑40% frame is sunset sky; gradient peach‑coral → blush‑pink → lavender, soft thin sunset clouds. Distant low California hills in dusty‑lavender / blue‑grey. NO buildings / landmarks.
Pool: large rounded‑rectangle drained pool, curved corners, sloped inner walls, flat dry clean floor. Late‑90s So‑Cal residential pool styling, polished fashion‑film location NOT abandoned ruin. Inner plaster warm off‑white with pale blush tones, thin dusty‑blue tile trim, small drain, minor stainless‑steel fixtures, optional small chrome ladder far side. Pool floor clean dry, NO water, puddles, graffiti, logos, mosaics.
Deck & landscape: narrow pale warm concrete deck. Keep open distant hills visible, no enclosing walls.
Lighting: soft natural sunset peach‑pink skylight, pale‑lavender shadows inside basin. NO artificial pool lights / spotlights / glows. Bright clean sunset exposure.
Style: early‑2000s pop‑rap MV aesthetic, subtle digital‑camera softness, fine analog grain, realistic plaster‑ceramic‑concrete textures, dreamy youthful mood.
Negative: NO humans, mannequins, personal belongings. NO dark sky / storm / night. NO indoor‑studio cyclorama / black enclosing walls. NO water, pool toys, furniture, resort decor, graffiti, damage. NO fisheye distortion, cropped pool perimeter. NO text, signs, logos, watermarks.
Video Prompt (Character):
15‑second, 16:9 horizontal, photorealistic live‑action fashion‑music short.
Asset locks:
Location:@f630634b‑ff1d‑4a8f‑93c8‑32617aad2d32
Character:@e5cb81dd‑5569‑4f18‑a8e8‑acdfea818226
Master audio‑timing‑vocal reference:@a9608cdb‑6b90‑47d0‑8a67‑de8ae14014d8
Rules Highest Priority B: ALL filming‑equipment (camera‑arm‑track‑crew‑light‑drone etc) 100% invisible. NO equipment reflection on glasses‑water‑metal props. NO visualized camera‑path‑rail‑HUD.
Camera physics: Continuous single‑take robotic‑arm S‑curve movement, real parallax for background. NO digital‑zoom / jump‑cuts. Only 1 allowed hidden stitch:7.55‑7.90s(hair crosses frame edge).
Audio master clock: strictly sync to @a9608cdb‑6b90‑47d0‑8a67‑de8ae14014d8, BPM126.1 4/4.
Vocal onset timestamps:0.52,1.46,2.42,3.36,4.34,5.26,6.24,7.18,8.14,9.08,10.04,10.98,11.94,12.88; final whisper ~14.05s. Max ±0.05s time offset. Decelerate 0.10‑0.14s before each vocal onset.
Location lock: Only the reference open‑air pool scene. Character stays dry concrete pool front‑left ≥1m from edge, NO entering‑water / sitting‑pool‑edge. Preserve all original architecture & distant landscape.
Character lock: keep reference character’s face‑age‑body‑hair‑sunglasses‑clothes‑accessories 100% consistent. Lips‑jaw‑cheeks move realistically for singing. Opening shot must show full‑body + pool environment.
Camera shot sequence (continuous robotic‑arm movement):
0.00‑0.52s｜High overhead wide shot, character full‑body in frame, lips closed pre‑singing.
0.52‑1.46s｜1st UNLIMITED: descend + push‑in → left‑face close‑up, sing.
1.46‑2.42s｜2nd UNLIMITED: hold left close‑up → translate & pull‑back.
2.42‑3.36s｜3rd UNLIMITED: pull‑back to waist‑up frontal medium shot, sing.
3.36‑4.34s｜4th UNLIMITED: hold frontal medium, slight weight‑shift, sing.
4.34‑5.26s｜5th UNLIMITED: push forward → right‑face close‑up; brief thin lens‑flare permitted here only.
5.26‑6.24s｜6th UNLIMITED: hold right‑face close‑up, sing.
6.24‑7.18s｜7th UNLIMITED: descend‑pull‑back‑orbit to low rear‑left medium‑wide.
7.18‑8.14s｜8th UNLIMITED：rear‑left shot; 7.55‑7.90s allowed hidden stitch; slight 70% hair‑body slow‑mo, lip‑sync stays real‑time.
8.14‑9.08s｜9th UNLIMITED：move to low front‑right medium shot, sing.
9.08‑10.04s｜10th UNLIMITED：finger lightly touch sunglass temple(DO NOT remove glasses), sing, hand falls.
10.04‑10.98s｜11th UNLIMITED：rise to eye‑level frontal close shot, sing.
10.98‑11.94s｜12th UNLIMITED：hold shoulders‑up frontal close‑up, sing.
11.94‑12.88s｜13th UNLIMITED：fast push‑in along optical‑axis.
12.88‑13.75s｜14th UNLIMITED：extreme lips close‑up, clear full articulation.
14.05‑15.00s｜15th whispered UNLIMITED: keep lips extreme‑CU; smaller but visible syllable articulation. Lips close; hard‑stop picture+audio exactly at 15.00s, NO fade‑out.
Lighting: unified late‑afternoon sun‑sky lighting; only allowed flare is during 4.34‑5.26s. Realistic skin‑water‑concrete material, natural reflections.
Audio rule: use master audio track exactly as‑is, DO NOT regenerate vocal. Only allow extremely subtle sea‑breeze / fabric / water ambient sounds; NO robotic‑drone‑motor‑shutter sound‑effects. NO extra vocals/lyrics.
Negative Constraints Summary:
No shut‑mouth fake singing; no visible production‑equipment/reflection; no digital‑zoom / jump‑cut / wrong parallax; no character‑face‑hair‑outfit drift; no dancing‑hip‑sway‑exaggerated‑gestures; no removing‑altering sunglasses; no falling‑into‑pool; no neon‑cyberpunk‑plastic‑skin‑over‑HDR; no template‑transition‑glitch‑freeze‑frame; NO subtitles‑text‑logo‑watermark. Final: continuous robotic‑arm camera flow; every vocal trigger visible lip‑jaw performance; equipment fully hidden.
Screenshot example
Want to recreate this look in minutes? Upload your references, paste the prompt, and generate your own music video with Pollo AI.