Prompt Hotel Lobby AI : texte à copier pour deux photos

Prompt Hotel Lobby AI

Create one continuous 13-second silent live-action duo performance featuring exactly the two subjects in the reference photos. Use the FIRST reference photo only for subject A on SCREEN LEFT and the SECOND reference photo only for subject B on SCREEN RIGHT. Assign roles by image order, never by gender or species: either subject may be a person or an animal. Preserve each subject's individual identity, face or muzzle, hair or fur, markings, proportions and any clothing actually present in their own photo. Do not exchange, blend or duplicate identities. Do not invent a specific gender, species, outfit or clothing color. Scene: a warm orange seamless recording studio, with one silver studio microphone suspended from the top center between the subjects. Medium-wide two-subject composition, with both faces, upper bodies and expressive limbs visible, enough room for a bent-limb gesture and a complete torso turn. A stays on screen left; B stays on screen right even while turning. Locked-off camera, natural studio lighting, realistic anatomy and skin or fur. No text, logos, subtitles, extra subjects or scene cuts. Performance: an energetic, playful rap-duo performance. Both make small rhythmic head and body movements to a steady brisk hip-hop pulse. A is the lead performer for most of the clip, with quick, varied visible mouth movements: distinct closing, reopening, wider and rounded shapes, in short clusters with brief pauses. Avoid a frozen open mouth, a sustained small-O shape, a constant smile or identical repeated jaw cycles. B mostly reacts through gestures and head movements rather than continuously mouthing words. Movements flow naturally between sections without resetting at timestamps. Adapt every action to each subject's existing anatomy. For animals, retain the original species, muzzle, fur and natural limbs; use small muzzle/jaw changes and expressive head, body or front-limb movements instead of human lips, hands or invented human limbs. A pointing gesture may become a natural front-paw accent. The dab may become a head dip toward a bent front limb; the shh ending may become a front paw near the muzzle if anatomically plausible, otherwise a quiet still pose facing inward. Never force an animal into an unnatural upright stance. Preserve recognizable intent and action order without distorting anatomy. Highest-priority actions: B's clear bent-elbow dab or anatomically suitable bent-limb head dip at 8.0–9.0 seconds, a complete torso turn showing the back at 9.0–10.5 seconds, then a smooth return toward A and the camera at 10.5–11.5 seconds. Use clear silhouettes and a brief back-facing beat, not just a sideways glance. A continues the lead articulation during this sequence. 0.0–3.0 seconds: A faces slightly inward toward the microphone and camera, leads with quick expressive mouth movements, small head nods and a relaxed gesture near the waist or chest. B extends one arm or suitable front limb forward, then gently pulses it downward. Around 2 seconds A makes a short forward point or natural limb accent, then lowers it. Neither subject touches or blocks the microphone. 3.0–5.5 seconds: A continues quick mouth articulation and a relaxed body bounce, briefly turning toward B. B lowers the extended limb, sweeps it diagonally downward where anatomically possible, then turns back toward A with a playful response. Keep positions and motion continuous. 5.5–8.0 seconds: A continues the lead articulation with alternating small gestures near the body. B presents an open hand or suitable front-limb response toward A, draws it back, then gives one cross-body accent if natural for the subject. B mostly keeps the mouth relaxed or closed. Do not switch to simultaneous continuous rapping. 8.0–9.0 seconds: B performs one clear energetic dab: for a human, bend one elbow across the face, dip the head into the inner elbow and extend the opposite arm diagonally toward screen right. For an animal, use the anatomically suitable head dip and bent-limb equivalent described above. Hold only briefly, then release; do not remain bowed for several seconds. A remains visible on screen left, continuing quick varied articulation and small rhythmic accents. 9.0–10.5 seconds: After releasing the gesture, B rotates the torso fully away until the back faces the camera clearly for a brief beat. B stays on screen right, gives a small outward limb or body accent and never crosses in front of A. A keeps facing the camera or microphone with quick varied articulation. 10.5–11.5 seconds: B smoothly turns back toward A and the camera, revealing the face again and lowering the limbs. Complete the return by 11.5 seconds. A continues the lead mouth movements until the shared ending. 11.5–13.0 seconds: Both stop the quick mouth movements and settle into a playful shh ending facing inward, with both faces visible in three-quarter view. For humans, raise one index finger near their own lips; adapt naturally for animals as described above. Hold briefly. End cleanly without a fade, scene change or extra action. Generate no audio, music, lyrics, captions or sound effects. The visible articulation is silent rhythmic performance, not a guarantee of word-level lip sync. Audio, when selected, is added separately.