{ "task": "Photorealistic garden conversation between Idris and…

{ "task": "Photorealistic garden conversation between Idris and Murad. Idris is about to send an angry reply; Murad asks him to wait.", "format": { "duration": "15 seconds", "aspect_ratio": "16:9", "shot": "One continuous take. No cuts or hidden transitions." }, "references": { "@image1": "Exact opening frame: Idris's hands, phone, unsent draft, brown sleeves, framing, light and colors. Murad and his chair are outside this crop, not absent from the scene.", "@image2": "Spatial master: preserve opposite seating, furniture, table orientation, distances, tableware, garden and human scale.", "@image3": "Idris identity and wardrobe: dark wavy hair, full beard without moustache, lean build, brown linen shirt, taupe trousers, brown shoes.", "@image4": "Murad identity and wardrobe: bald, full beard without moustache, sturdy natural build, olive overshirt, ivory T-shirt, beige trousers, brown shoes.", "exclusions": "Do not copy studio backgrounds, studio lighting or character-sheet layouts.", "@audio1": "Idris, first line.", "@audio2": "Murad, second line.", "@audio3": "Idris, third line." }, "priorities": [ "Assign each audio only to its specified speaker.", "Keep both men seated opposite each other. Camera movement never rearranges the scene." ], "audio": { "source": "Use each uploaded recording for its assigned line. Preserve voice, Russian pronunciation, emotion, breathing and natural pace.", "order": "@audio1, then @audio2, then @audio3. Each once, without overlap.", "language": "Native Russian. Latin-letter dialogue is a phonetic guide, not English or on-screen text.", "timing": "Preserve internal pauses and original playback speed. Use remaining time for reactions. Do not accelerate, stretch or truncate speech.", "lip_sync": "Match the visible speaker's lips and jaw to his recording. The listener reacts silently.", "exclude": "No extra dialogue, replacement voices, repetition, narrator or later lines about a tree." }, "camera": { "start": "Match @image1: beside and behind Idris's right shoulder, looking down at his hands and phone.", "move": "Pull back smoothly to clear his shoulder, then arc gently outside the adjacent table corner toward the open long side. Gradually rise to seated eye level.", "reveal": "Reveal Idris's three-quarter face during his first line. Shift focus from phone to face. Widen to include Murad across the table.", "arrival": "Reach a medium two-shot by 10 seconds, with the tabletop separating the men. Both faces remain readable.", "end": "Ease into a stable composition and hold for continuation.", "geometry": "Stay on one side of the dialogue axis. Clear route outside bodies and furniture. Preserve eyelines.", "optics": "Natural perspective, fixed focal length, smooth focus adjustment.", "exclude": "No digital zoom, whip pan, full orbit, mirrored view, teleportation or passage through objects." }, "timeline": [ { "time": "0-5s", "speaker": "Idris", "audio": "@audio1", "dialogue": "Ty slyshal, chto on pri vsekh skazal?", "emotion": "Hurt beneath restrained indignation; seeking acknowledgment, no shouting.", "action": "Right thumb hovers above send. Left hand supports the phone; right steadies its edge. After a brief hesitation, he speaks and looks toward Murad.", "camera": "Move from the phone toward Idris's three-quarter face." }, { "time": "5-8s", "speaker": "Murad", "audio": "@audio2", "dialogue": "Slyshal. Ne speshi otvechat'.", "emotion": "Understanding, calm and gently firm; no lecturing.", "action": "Murad maintains attentive eye contact. Idris withdraws his thumb from send to the phone's edge.", "camera": "Continue revealing Murad opposite Idris." }, { "time": "8-9s", "speaker": "None", "action": "Idris takes a small breath, absorbing the reply before objecting. He keeps holding the phone.", "camera": "Continue gently toward the two-shot." }, { "time": "9-13s", "speaker": "Idris", "audio": "@audio3", "dialogue": "Ya chto, dolzhen eto tak ostavit'?", "emotion": "Defensive frustration, controlled intensity, seeking support.", "action": "Small forward lean and raised eyebrows. Phone stays low in his hands. Murad listens without interrupting.", "camera": "Settle into the medium two-shot." }, { "time": "13-15s", "speaker": "None", "action": "Murad holds thoughtful eye contact, preparing to answer. Idris waits. Natural breathing and blinking, no freeze.", "camera": "Hold the two-shot. No fade or concluding gesture." } ], "continuity": { "count": "Two adult men, two original chairs, one table, one phone and two tea glasses with saucers.", "seating": "Original opposite positions. Hips supported, feet on grass, knees beneath the table. No furniture intersections.", "scale": "Natural adult proportions. Murad is sturdier than Idris, never oversized.", "props": "Preserve basket, plates, runner and teapot positions. No duplication or rearrangement.", "phone": "Idris holds the same phone throughout. Draft remains unsent; preserve the interface. No notifications or new text.", "identity": "Preserve faces, clothing, hair and beards without moustaches.", "anatomy": "Coherent hands, fingers, wrists and supported contact.", "acting": "Natural breathing, blinking and small unsynchronized reactions. Neither looks at the camera.", "story": "No pocket reach yet. No oud, burner, charcoal, flame or smoke." }, "look": { "light": "Preserve source afternoon sunlight through foliage and soft shaded fill.", "color": "Match reference exposure, white balance, saturation, contrast and photographic texture. No new grade.", "materials": "Natural skin, linen, cotton, wood grain, restrained reflections and contact shadows.", "background": "Subtle foliage movement only." }, "sound": { "dialogue": "The three assigned recordings, clear and foregrounded.", "ambience": "Very quiet leaves and distant birds below speech.", "foley": "Subtle fabric and contact sounds motivated by visible action.", "music": "NONE: no instruments, singing, humming, choir, nasheed, melodic drones or rhythmic backing.", "exclude": "Notifications, impacts, whooshes and narration." }, "end_state": "Stable medium two-shot. Idris holds the unsent phone low in both hands and looks at Murad. Murad looks back, hands relaxed, ready to answer. Preserve this state for continuation.", "negative": "Speaker swap, doubled speech, overlapping audio, clipped words, listener lip-syncing, English pronunciation, music, subtitles, captions, cuts, mirrored geography, men sitting side by side, moved furniture, identity drift, extra people or limbs, malformed hands, giant bodies, floating phone, sent message, duplicated props, alcohol, smoke, plastic skin, HDR, oversharpening or changed reference colors." }