KI-Video-Prompt-Generator für Sora, Veo 3 und Universal

- Sora Prompt A slow dolly-in through a rain-soaked neon alley at night, 35mm lens, shallow depth of field. The subject, a lone figure in a wet trench coat, pauses under a flickering sodium lamp, exhaling visible breath. Reflections of magenta and cyan signage ripple in puddles. Camera glides at chest height, maintaining 1.5 feet per second, with subtle handheld micro-jitter. Lighting: practical neon sources, key light from a distant storefront, rim light from a passing taxi. Atmosphere: light fog, drizzle, no wind. Motion: figure turns head slightly, then walks left out of frame, leaving empty alley for 2 seconds. Aspect ratio 16:9, 24fps, color grade teal-orange, grain 35mm. - Gemini Veo 3 Prompt A moody, dreamlike sequence of a solitary dancer in an abandoned industrial hall, dust motes floating in shafts of golden hour light. Slow-motion 60fps, lens flare from a high window, warm amber and deep shadow contrast. The dancer moves in a slow, improvised spiral, fabric trailing behind, motion blur on edges. Camera orbits at a low angle, 10 degrees per second, with a slight push-in. Sound design implied: distant hum, soft footsteps, no music. Emotional tone: melancholic, intimate, timeless. Lighting: single directional sunbeam, bounce fill from concrete floor. No cuts, one continuous take, 8 seconds. Aspect ratio 2.39:1, color palette muted ochre and charcoal. - Universal Prompt A cinematic wide shot of a small boat drifting on a calm, misty lake at dawn. Camera starts at water level, slowly rising to a 45-degree angle over 6 seconds. The boat rocks gently, ripples spreading outward. A single heron stands on the bow, then lifts off, wings catching pale pink light. Composition: rule of thirds, boat on left, open water on right. Lighting: soft diffused sunrise, low contrast, pastel sky. Movement: static camera with a vertical tilt, no zoom. Atmosphere: quiet, serene, slightly surreal. Duration 8 seconds, 24fps, natural color grade, no text or overlays. Works with any text-to-video model.