Cat and Marmot Car singing VIDEO Breakdown (Prompts Included)
Tutorial：
Open pollo.ai, head to Video Generation. Upload the referencce image or pick assets from your Creations library. Select reference-to-video mode and paste your video prompt. You can also upload a video for motion reference.
Note: Add @your-image for image reference and @your-video for video reference inside the prompt. Pick any of these models: Seedance 2.5, Seedance 2.0, MiniMax H3 or Wan 3.0. Click Generate after confirming all settings.
Sticker 1
Sticker 2
Video Prompt
CAR, LIGHTING, AND CAMERA CONTINUITY:
Use one believable compact crossover SUV. The cabin has black front and rear headrests, black seat fabric, a light-gray headliner, clear side windows, ordinary seatbelt anchors, and no visible manufacturer badge. Daylight enters softly through the windshield and side windows. Outside, show gently passing sidewalks, low-rise storefronts or apartments, street trees, and distant parked cars, all without readable signs, logos, license plates, or recognizable locations.
The primary camera is a stable wide dashboard-mounted camera looking back at the two front seats. It frames CAT at screen-left and MARMOT at screen-right, with the rear seats centered behind them and a small portion of the steering wheel in MARMOT's foreground. Supplement it with clean interior close-ups from the dashboard or passenger-side area. These are edited music-video cuts, not one continuous take. Preserve the same seat positions, car interior, belt direction, lighting direction, and road travel across every cut.
CONTINUOUS PERFORMANCE AND EDIT PLAN:
0.0-3.5 seconds - Duet begins in the dashboard wide:
Open immediately on the stable, centered dashboard view. CAT sits upright in the front passenger seat at screen-left, MARMOT sits upright in the driver's seat at screen-right, and both are visibly belted. They enter the first beat together with exaggerated but charming lip-sync: CAT gives one small outward forepaw flick toward the lens, while MARMOT bobs its head and opens its mouth broadly on the first sustained "ah-oh." Keep the rear seats centered in the background, road scenery flowing gently through the windows, and the steering wheel visibly connected to MARMOT's position.
3.5-6.5 seconds - MARMOT's lead close-up:
Cut to a tight dashboard-side close-up of MARMOT from the chest up. Its seatbelt is clear across the brown-tan chest, with the driver-side window and moving street softly behind it. MARMOT performs the lead phrase with a comically huge open mouth, visible incisors, a slight head tilt, and one forepaw lifted near the lens as though marking the beat. The raised paw never blocks the face for more than a moment. Keep its gray face, snout, eyes, incisors, and fur identity stable.
6.5-9.5 seconds - CAT answer close-up:
Cut to a close passenger-side view of CAT. CAT leans slightly forward within the seatbelt, closes its eyes for one short emphatic note, opens its mouth on the answer vocal, and pumps one compact forepaw toward the camera in time. Hold its folded ears, round orange-striped face, green eyes, pale muzzle, pink nose, and white whiskers sharply readable. The passenger headrest and soft street movement remain behind it; do not turn CAT into a human singer or add a microphone.
9.5-13.0 seconds - Back-and-forth two-shot:
Return to the dashboard two-shot. CAT looks toward MARMOT and performs two gentle alternating forepaw motions, as if playing an invisible rhythm in the air. MARMOT looks ahead, then briefly glances toward CAT while continuing its wide-mouth lip-sync. Both bounce subtly in their seats to the beat without leaving their belts. The car continues straight and calm; no sudden steering, braking, swerving, or dangerous road action.
13.0-16.0 seconds - CAT's dramatic mini-solo:
Cut to CAT's close-up again, slightly lower and more energetic. CAT opens its mouth for a longer sustained vowel, narrows its green eyes with mock intensity, then raises one forepaw near its cheek and points it outward. Its whiskers and folded ears remain stable. End the shot with CAT's forepaw lowering naturally back toward its lap.
16.0-19.0 seconds - MARMOT's response:
Cut to MARMOT in a front three-quarter close-up. MARMOT sings the response with a big cheerful open-mouth expression, makes one short raised-paw accent, then returns that paw close to the steering-wheel area. The seatbelt remains correctly placed. The camera captures the gray muzzle, visible incisors, dark eyes, and the chunky dark paw without distortion or extra digits.
19.0-22.5 seconds - Big shared chorus:
Return to the dashboard wide for the strongest shared refrain. CAT and MARMOT sing at the same time, mouth movements synchronized to the original "ah-oh" chorus. CAT performs a compact double forepaw groove without crossing into MARMOT's space. MARMOT nods to the rhythm and alternates between looking forward and giving CAT a quick approving glance. Keep their expressions joyful, their bodies seated, their belts fastened, and all car geometry stable.
22.5-25.5 seconds - CAT turns to the passing world:
Cut to a slightly wider passenger-side interior angle. CAT turns its round orange head toward the passenger-side window and holds one delighted sustained note, watching the sunny roadside slide past. MARMOT remains in the background at screen-right, still singing and gesturing lightly. CAT's body stays seated and restrained by the belt; it does not lean out of the window or leave the seat.
25.5-26.7 seconds - Brief window-view release:
Cut briefly to a passenger-side exterior-facing window view from inside the moving vehicle. Show an unbranded beige-and-brick low-rise streetscape, a leafy curbside tree, a red-painted curb, a single distant cyclist, and smooth lateral motion. The lower edge may contain a small blurred part of the car's side mirror or window frame. This is only a quick musical breath; show no readable store signs, license plates, logos, or text.
26.7-30.0 seconds - Final driver-side payoff:
Cut back inside to a passenger-side three-quarter shot focused on MARMOT at screen-right, with CAT visible behind and to the left in the passenger seat. MARMOT takes the final big original vocal syllable, raising one forepaw in a confident final beat while the other remains near the steering wheel. CAT turns back from the window, joins the final "oh," and settles into a proud, deadpan expression. End on MARMOT's open-mouth smile and raised paw, with both seatbelts visible and the sunny street still drifting past the windows.
VISUAL AND MOTION RULES:
Photorealistic live-action animal comedy with detailed short cat fur, distinct tabby striping, stable folded ears and green eyes, plus detailed marmot fur, gray muzzle, stable incisors, and realistic chunky paws. The performance is intentionally anthropomorphic only in the limited karaoke gestures and lip-sync expressions; both animals remain seated animals with believable body mass and proportions. Preserve exactly one CAT, one MARMOT, one SUV cabin, and one dark seatbelt on each animal. Every visible forepaw must have correct natural anatomy. Do not add a third animal, a human driver, human passengers, microphones, instruments, costumes, or extra props.
Use natural daytime exposure, modest shallow depth of field in close-ups, ordinary consumer-camera clarity, slight realistic vehicle vibration, and restrained motion blur outside the windows. Maintain a bright, comic, friendly tone. Cuts should land on musical phrases, but no shot should look like a morph, teleportation, or unrelated car.
AUDIO:
Create an original bright, bouncy pop-karaoke backing track with a simple drum beat, soft bass, handclap-like percussion, and a short non-identifying wordless vocal hook built from "ah-oh" and "oh-yeah" syllables. Synchronize CAT's and MARMOT's mouth movements to their alternating and shared sung syllables. Add subtle cabin tone, gentle road rumble, light tire noise, and a faint passing-street ambience under the music. Keep the music louder than the road sound. No recognizable song, lyric quotation, spoken dialogue, narrator, radio announcer, captions, subtitles, animal noises, horns, sirens, or sudden sound effects.
NEGATIVE PROMPT:
Extra cat, extra marmot, duplicate animal, missing animal, kitten, different cat breed, unfolded cat ears, wrong cat coat color, changed cat eye color, missing whiskers, changed cat face, different rodent, beaver, guinea pig, capybara, wrong marmot face, missing marmot incisors, extra incisors, changed marmot fur color, human in animal costume, human hands, human face, animal clothing, collar, glasses, headphones, hats, microphone, instrument, food, drink, phone, props, CAT driving, CAT in driver seat, MARMOT in passenger seat, seat swap, unbuckled seatbelt, belt crossing wrong direction, belt passing through body, floating steering wheel, disconnected steering wheel, vehicle-brand logo, readable signs, readable license plates, captions, subtitles, lyrics on screen, watermark, text, distorted paws, extra paws, extra digits, broken teeth, warped mouth, face drift, identity drift, animal standing on seat, animal leaving seat, animal leaning out window, dangerous driving, sharp turn, braking, crash, road hazard, crowd, dark night lighting, horror tone, cartoon render, animation, excessive camera shake, incoherent cuts, morph transition, low resolution, blurred face, recognizable song, copyrighted lyrics, dialogue, narration, barking, meowing, squeaking, loud horns, sirens, audio mismatch.
Final Summary
Create one complete 30-second horizontal 16:9 1080p photorealistic car-karaoke comedy at 30 fps. Use sticker1 as the fixed identity source for exactly one orange folded-ear tabby CAT in the front passenger seat, use sticker2 as the fixed identity source for exactly one gray-brown MARMOT in the driver's seat, and use Reference Video only for broad dashboard-camera energy, alternating close-up rhythm, sunny driving ambience, and comic musical timing.
The finished video begins with a dashboard two-shot, cuts between CAT's compact forepaw-led vocal answers and MARMOT's open-mouth lead vocal, builds into a shared front-seat chorus, briefly turns CAT toward the passenger window, uses one short exterior-facing window view, and ends on MARMOT's raised-paw final note. Preserve both animals' source identities, screen positions, correctly fastened seatbelts, the same black SUV interior, clear original non-identifying music, calm road motion, no text, no extra characters, no recognizable song, and no watermark.