社群 AI 提示詞

Transform the input[0] into a high-quality 2D Japanese anime-style video. Preserve the original motion exactly as it appears in the source video, including the full choreography, motion structure, rhythm, body weight shifts, limb trajectories, hand gestures, head movements, turns, footsteps, performance intensity, and camera composition. Do not alter the cho

Convert the input[0] into a rotoscoping-style storyboard line-art motion reference video. Preserve the original movement structure, rhythm, limb trajectories, body weight shifts, and pose transitions from the source video as accurately as possible. Do not redesign the choreography or add any extra performance actions. The character should be rendered in a ro

Convert the live-action input[0] into a de-characterized 3D mannequin motion reference video. The final result should preserve only the performer’s motion structure, pose changes, rhythm, weight shifts, and limb trajectories, while matching the original movement as closely as possible. Keep the choreography exactly the same as the source video, without chang

Transform the real human dancer in the input video into a 2D anime character in the style of JoJo's Bizarre Adventure ref[0]. Preserve the original dance choreography, motion timing, body movement, camera angle, character position, rhythm, and video duration exactly. Only change the visual style. The character should have a strong JoJo anime look: sculptura

[Quality and Technical Specs] Generate a 5s, 16:9 refs_generate AI video template titled "Live2D Desktop Companion Idle". Use polished anime Live2D desktop companion style with visual novel UI, dark translucent glass panels, warm room light, blue screen glow, clean line art, soft color blocks, and a cozy interactive mood. [Hard Settings] input[0] is the sol

[Quality and Technical Specs] Generate a 5s, 9:16 refs_generate AI video template titled "Low-Angle Prince Presence". Use anime Live2D character animation with a strong low-angle perspective, black formal suit or butler-style clothing, cold confident expression, dark minimal background, cool blue shadows, silver rim light, and refined premium polish. [Hard

[Quality and Technical Specs] Generate a 10s, 9:16 refs_generate AI video template titled "Fox Mask Cool Reveal". Use refined semi-realistic anime Live2D animation with Korean manhwa polish, otome game CG elegance, cold white lighting, silver hair detail, black glossy clothing, ice blue shadows, pale violet rim light, and a mysterious restrained mood. [Hard

[Quality and Technical Specs] Generate a 5s, 1:1 refs_generate AI video template titled "Live2D Fixed Portrait Micro. Use high-definition Live2D portrait animation with 8K-level facial clarity, stable facial features, about 40 percent skin filtering while preserving natural texture, dynamic iris highlight tracking, soft white light, natural skin tones, subtl

[Quality and Technical Specs] Generate a 10s, 9:16 refs_generate AI video template that turns the character in input[0] into a high-quality Live2D animated portrait. The style is clean, delicate, polished, social-video-ready dynamic illustration with soft white, pale pink, light gold, cyan blue, and soft violet color accents. [Hard Settings] input[0] is the

[Image Objective] Generate one premium 2:3 vertical male VTuber Live2D model bundle character card for an original adult man named "ASTER VALE". The result must be a complete character design board, not a single-person illustration or poster. Use a refined white archive composition for a professional Live2D-ready character sheet. [Character Concept] Create

[Image Objective] Generate one premium 2:3 vertical male VTuber model bundle character card for an original adult man named "NOCTIS VELLUM". The output must be a multi-zone production-ready Live2D character design sheet, not a single illustration, portrait, ordinary poster, 3D CG render, physical doll photograph, or plastic toy. [Character Concept] Create a

[Image Objective] Generate one premium 16:9 horizontal character sheet for an original adult female VTuber named "AURORA NULL". The result must be a professional, model-ready Live2D design sheet with refined 2.5D depth, not a real doll photograph, toy render, flat anime poster, or 3D CG render. [Core Character and Materials] Create a BJD-inspired adult fema

A tiny dragon visits a castle kitchen. A nervous chef offers dinner, then discovers the visitor prefers dessert. Warm firelight, expressive reactions, a playful cinematic camera move.

Create a 10-second cinematic first-person POV video from inside a refrigerator looking outward, as if the camera is hidden behind cold shelves, glass bottles, and food containers. Keep the fridge frame, bottles, and containers close to the lens as foreground depth. At 0-2s, a sharply dressed man opens the refrigerator door. Cold white fridge light floods hi

Create a 10-second cinematic 2D American comic animation from a first-person POV inside a refrigerator looking outward. The camera is hidden behind cold shelves, milk cartons, glass bottles, and snack boxes, making the viewer feel physically inside the fridge. Use input[0] as Character 1, a stylish teenage boy. Use input[1] as Character 2, a stylish teenage

Create an 8-second cinematic Korean neo-noir apartment suspense video from a true peephole POV, using input[0] as the visual reference for the visitor, wardrobe, hallway style, and color palette. The camera is inside an apartment door looking outward through a round door peephole. Keep a subtle circular peephole feeling throughout: soft dark round vignette a

Create a 7-second cinematic black-comedy crime video from a true trash can POV in 16:9 widescreen, using input[0] as the visual reference for character design, wardrobe, props, alley mood, and color palette. The camera is fixed near the bottom inside a street trash can, looking upward through the open rim. Keep the dirty trash can rim and pieces of garbage c

Cinematic live-action video, about 10 seconds. Static POV from inside the open trunk of a car, looking upward and outward through the trunk opening. The trunk interior creates a dark curved frame around the image, with the rubber seal and side panels visible at the edges. Nighttime roadside police inspection, red and blue patrol lights flashing softly in the

Cinematic live-action ocean thriller video, 16:9 landscape, about 10 seconds. Underwater predator POV from below the ocean surface, looking upward toward the bright moving waterline. The camera represents a shark-like predator moving through the water, but the animal itself is never visible. The surface above shimmers with sunlight, waves, bubbles, and suspe

2D animated dark fairytale horror video, 16:9 landscape, about 8-10 seconds. Static POV from underneath a bed, looking outward from the floor-level space beneath the bed. The camera remains fixed under the bed for the entire shot. The left and right sides of the frame show bed legs and the lower edges of the bed frame. The edge of the bedsheet hangs down nea

Transform the real human dancer in the input video into a 2D anime character in the style of JoJo's Bizarre Adventure ref[0]. Preserve the original dance choreography, motion timing, body movement, camera angle, character position, rhythm, and video duration exactly. Only change the visual style. The character should have a strong JoJo anime look: sculptura

input[0] is the character reference image. Strictly use input[0] as the primary visual identity reference for the protagonist: extract and maintain the same person's face, skin tone, hairstyle, body type, costume, accessories, aura, and overall style. 【Visual Style】 1990s Japanese hand-drawn cel animation aesthetic (retro 90s cel animation), in the art style

[Quality and Technical Specs] Create a 15s, 16:9 feature-film 2D anime opening with crisp cel shading, Pop Art manga graphics, and a red/black/white phantom-thief palette accented by yellow and cool neon blue. Keep the action fast, theatrical, legible, and spatially coherent. [Hard Settings] input[0] is the sole identity reference for the visible third-pers

input[0] is the character reference image. Strictly use input[0] as the primary visual identity reference for the protagonist: extract and maintain the same person's face, skin tone, hairstyle, body type, costume, accessories, aura, and overall style. 【Overall Style】 TV anime-style cel shading rendering, Ume Aoki-style character design (rounded soft facial f

input[0] is the character reference image. Strictly use input[0] as the primary visual identity reference for the protagonist: extract and maintain the same person's face, skin tone, hairstyle, body type, costume, accessories, aura, and overall style. Do not show the reference image itself. Do not generate a character sheet. [Overall Style] Top-tier Japanes

[Quality and Technical Specs] Create a 15s, 16:9 premium 2D cel-animated street-action opening with crisp 3D-to-2D shading, elastic squash-and-stretch, high-saturation electric blue/green/pink/orange/violet color clashes, rhythmic camera motion, and clearly readable action. [Hard Settings] input[0] is the sole identity reference for the visible third-person

input[0] is the character reference image. Strictly use input[0] as the primary visual identity reference for the protagonist: extract and maintain the same person's face, skin tone, hairstyle, body type, costume, accessories, aura, and overall style. Do not show the reference image itself. Do not generate a character sheet. Cinematic CG anime action sequenc

input[0] is the character reference image. Strictly use input[0] as the visual identity reference for the protagonist: extract and maintain the same person's face, skin tone, hairstyle, body type, clothing, accessories, demeanor, and overall style. Do not show the reference image itself. Do not generate a character sheet. One-shot long take, Michael Bay-styl

[Quality and Technical Specs] Generate a 10s, 16:9 refs_generate AI video template for Ink Bamboo Dual Blade Spear Duel. Style: dark wuxia Chinese animation with ink-wash bamboo, silver-white and ink-green contrast, and fast fight rhythm. The video must be polished, stable, rhythm-driven, visually coherent, and suitable for reusable template generation. [Ha

第一人称乙女向恋爱氛围视频，真人电影质感，参考图生视频。input[0] 是男主形象参考图，不是主角。请严格把 input[0] 作为画面中男主的身份参考，提取同一个人的脸、肤色、发型、身形、服装、配饰、气质和整体风格；不要在画面中展示参考图本身，不要生成多个人物设定板，不要让参考图人物变成第一人称主角。 镜头设定：镜头就是主角的第一人称视角，主角本人完全不出镜。不要出现主角的脸、身体、手、手臂、脚、衣服、影子或倒影。观众只能通过镜头感受到自己正在被男主注视和欢迎。 男主设定：男主必须出现在镜头前，作为欢迎主角的人。男主的长相、发型、肤色、服装、配饰、气质和细节应尽量与 input[0] 保持一致。男主是英式魔法学院风格的学生或青年角色，穿着带有魔法学院感的深色制服、长袍或外套，具有精致服装层次、皮革或金属

[Quality and Technical Specs] Create a 15s, 16:9 cinematic desert mystery-fantasy adventure with realistic PBR materials, HDR sunset light, drifting sand, ancient runes, futuristic geometry, and a solemn gold/ochre/cool-blue/white palette. [Hard Settings] input[0] is the sole identity reference for the visible third-person hooded protagonist. Preserve face,

[Quality and Technical Specs] Create a 15s cinematic cyberpunk-wasteland action opening with ancient Chinese styling, realistic PBR materials, HDR lightning, teal/orange and dark-red neon, film grain, handheld urgency, and clear high-speed pursuit. [Hard Settings] input[0] is the sole identity reference for the visible third-person female hunter. Preserve f

input[0] is the character reference image. Strictly use input[0] as the primary visual identity reference for the protagonist: extract and maintain the same person's face, skin tone, hairstyle, body type, costume, accessories, aura, and overall style. 15-second 4K 60fps Hollywood-grade game CG skill animation. Forest Shrine Maiden. Mixed warm-and-cool color

Cyber assassin FPV game CG. 15-second 4K60fps Hollywood-grade game CG skill animation. Cyber assassin, neon cold color palette (blue/pink-purple), ultimate lightning effects, neon electric arcs surrounding. FPV first-person camera, ultra-smooth motion, no clipping, no text. 0-3s: FPV skims at high speed just above a rain-soaked cyberpunk street, neon reflect

[Quality and Technical Specs] Create a 15s, 16:9 premium cel-shaded 3D-to-2D ultimate-skill showcase with luminous translucent materials, clean linework, HDR starlight, crystal refraction, and a sacred, compassionate, magnificent mood. [Hard Settings] input[0] is the sole identity reference for the visible third-person protagonist. Preserve face, skin, hair

[Quality and Technical Specs] Create a 15s, 16:9 premium Chinese fantasy CGI sequence with high-saturation cel-shaded 3D, cinematic HDR light, and sophisticated black ink, gold leaf, dark-cyan energy, luminous flora, fluid simulation, and scroll closure. [Hard Settings] input[0] is the sole identity reference for the visible third-person Ink Immortal. Prese

[Quality and Technical Specs] Create a 15s, 16:9 realistic AAA open-world game demo set in brightly lit, rain-wet Roppongi at night. Use cinematic PBR materials, ray-traced neon reflections, premium glass UI, realistic gameplay camera lag, and a complete playable flow rather than a mobile-game advertisement. [Hard Settings] input[0] is the sole identity ref

input[0] is the character reference image. Strictly use input[0] as the visual identity reference for the main character: extract and maintain the same face, skin tone, hairstyle, body type, clothing, accessories, aura and overall aesthetic. Do not show the reference image itself. Do not generate a character reference sheet. Do not turn the reference charact

input[0] is the character reference image. Strictly use input[0] as the visual identity reference for the main character: extract and maintain the same face, skin tone, hairstyle, body type, clothing, accessories, aura and overall aesthetic. Do not show the reference image itself. Do not generate a character reference sheet. Do not turn the reference charact

Create a 15-second 9:16 vertical live-action photorealistic TikTok dance cover featuring two real-looking young female K-pop dancers, not anime, not illustration, not 3D cartoon. The video must look like a real smartphone recording of two human performers outdoors on a grassy slope, inspired by a KIRI X MINAMI ILLIT “It’s me” dance cover. Location: Olympic P

Create a 15-second 9:16 vertical live-action photorealistic AI K-pop girl-group dance video inspired by “Seasick Love Dance,” originally “Love Attack” by Rescene. The video should look like real young female idol performers filmed in a polished TikTok performance clip, not anime, not cartoon, not 3D doll animation. Scene: bright seaside-pop idol concept, cle

[Quality and Technical Specs] Generate a 10s, 9:16 refs_generate AI video template for Neon Rooftop Twin Dance. Style: sharp commercial dance video with neon rooftop contrast and synchronized movement. The video must be polished, stable, rhythm-driven, visually coherent, and suitable for reusable template generation. [Hard Settings] input[0] is the characte

[Quality and Technical Specs] Create a 15s cinematic 3D Chinese classical dance film with premium silk simulation, PBR jade and water, ray-traced moonlight, HDR detail, and a poetic, melancholic, elegant, serene mood in silver-blue, violet, and pearl-white. [Hard Settings] input[0] is the sole identity reference for the visible third-person dancer. Preserve

[Quality and Technical Specs] Generate a 10s, 16:9 AI video template. Visual style: cute realistic pet dance video with soft indoor daylight, plush pajama textures, clear fur detail, and playful synchronized movement. The output must be polished, stable, social-video-ready, and visually coherent. [Hard Settings] No input reference image is required. Generat

[Quality and Technical Specs] Generate a 10s, 9:16 AI video template. Visual style: high-saturation street dance video with a sweet cool idol mood, fixed mid-shot framing, soft street lighting, and crisp beat-driven motion. The output must be polished, stable, social-video-ready, and visually coherent. [Hard Settings] input[0] is the character reference ima

A 16:9 horizontal color IMAX film still during a blue-hour summer storm, with a realistic documentary feeling. A woman in her early twenties runs quickly across an empty straight rural road from the right side of the frame rather than toward the camera. She wears a loose pale-grey shirt and dark trousers, both damp with sweat and the first scattered raindro

A 16:9 horizontal color IMAX coming-of-age film still with a realistic documentary feeling. The first rain has begun at dusk on a narrow road along a river embankment. Wind presses down the tall grass on one side, while a dark river occupies the other. Two adults in their early twenties move along the same road in opposite directions. In the left foreground

A 16:9 horizontal color IMAX coming-of-age film still with a realistic documentary feeling. On a suffocating summer afternoon before a storm, an abandoned outdoor swimming pool is surrounded by tall neglected grass. The empty pool floor is dry and cracked, and the remaining white and pale-blue tiles are faded and yellowed. A man in his early twenties stands

Dystopian luxury advertising poster, bold red monochrome graphic design, glossy inflatable 3D typography. Central subject: hyper realistic Caucasian female face, slightly angled, uncanny symmetry, flawless skin with visible pores, enhanced cheekbones and lips, sealed in a transparent vacuum packed plastic bag like premium supermarket meat - tight plastic fol

A 16:9 horizontal color IMAX film still, captured as an accidental, incomplete moment from a live-action coming-of-age movie，style raw 100%. On a suffocating summer evening just before a storm, a faded old car is parked beside an empty rural service road. Two women in their early twenties occupy different depths of the frame. In the right foreground, a shor

[Quality and Technical Specs] Generate one high-quality Seedream 5.0 Lite image template. The result must be polished, visually coherent, detail-rich, and suitable for reusable CMS template generation. [Hard Settings] Generate the image directly from the text prompt. Do not require a user-uploaded reference image. Render a finished image, not a process sket

A photorealistic nighttime Seoul street hip-hop portrait containing exactly one clearly recognizable main subject: a 23-year-old adult East Asian man. He has natural Korean facial features, a defined jawline, short black hair, and realistic unretouched skin texture. He wears a black-and-white abstract-pattern baseball cap facing forward, with no real team lo

[Quality and Technical Specs] Generate one high-quality Seedream 5.0 Lite image template. The result must be polished, visually coherent, detail-rich, and suitable for reusable CMS template generation. [Hard Settings] input[0] is the sole visual identity reference. Preserve the referenced subject, costume, accessories, and defining traits; never show the re

Create one complete 15-second 9:16 vertical video in a single generation. The entire output must look like a raw smartphone screen recording captured directly from the phone operating system. The digital screen must fill the entire video frame edge to edge. Never show a physical phone, phone body, bezel, metal frame, notch, rounded device mockup, hand hold

Create one complete 15-second 9:16 vertical video as a flat, front-facing, edge-to-edge digital call interface. The interface itself must fill the entire canvas, like a direct screen capture with no depth, no perspective, and no surrounding object. Match the attached reference image only as a UI layout reference: dark burgundy-plum gradient background, cent

Generate a single, complete 15-second 9:16 vertical video. The entire output must look like a digital screen recording captured directly from a mobile phone operating system, with screen pixels fully filling the video frame from all four edges. No physical phones, phone bodies, bezels, metal edges, notches, rounded phone mockups, desktops, hands, room enviro

[Quality and Technical Specs] MiniMax H3, 720p, 15s, 16:9, ultra-real ocean monster cinema with dark desaturated grading, epic scale, plausible water displacement, heavy spray, and high-end but restrained visual effects. [Hard Settings] Text-to-video only. Keep one continuous world and the described subjects stable. [Scene] A cold gray storm sea with fog,

[Quality and Technical Specs] MiniMax H3, 720p, 15s, 16:9, large-scale ultra-real science fiction with real spatial depth, precise industrial surfaces, gradual light loss, controlled high-speed motion, and no visible pilot face. [Hard Settings] Text-to-video only. Keep one continuous world and the described subjects stable. [Scene] The interior of a coloss

[Quality and Technical Specs] MiniMax H3, 720p, 15s, 16:9, photoreal practical sci-fi effects with precise black metal materials, visible engineering seams, controlled industrial haze, and restrained dramatic lighting. [Hard Settings] Text-to-video only. Keep one continuous world and the described subjects stable. [Scene] A dark industrial hall with a scra

[Quality and Technical Specs] MiniMax H3, 720p, 15s, 16:9, ultra-real maritime adventure cinema with wet wood grain, believable metal armor, wind-driven mist, shallow depth of field, and strong but controlled camera shake. [Hard Settings] Use input[0] as the visual identity reference; retain visible traits and do not show the reference image. [Scene] A rai

[Quality and Technical Specs] MiniMax H3, 720p, 15s, 16:9, early-2000s consumer DV documentary footage with modest sensor noise, slight exposure hunting, soft interlaced texture, and authentic hand-held movement. [Hard Settings] Text-to-video only. Keep one continuous world and the described subjects stable. [Scene] The wooden deck of a late-medieval saili

[Quality and Technical Specs] MiniMax H3, 720p, 15s, 16:9, photorealistic fantasy disaster cinema with dramatic orange and yellow illumination, dark cloud texture, wet reflective surfaces, and restrained high-end visual effects. [Hard Settings] Text-to-video only. Keep one continuous world and the described subjects stable. [Scene] A turbulent open sea und

integrated_multimodal_description: Create a 15-second 16:9 audiovisual disaster-action sequence with photoreal live-action cinematography and high-end creature VFX. It is late afternoon in a dense modern city avenue destroyed by a giant kraken attack. Damaged towers, exposed steel, collapsed concrete, burning unmarked utility vehicles, drifting ash, backlit

integrated_multimodal_description: Create a 15-second 16:9 audiovisual game promo in pure 2D Japanese cel animation fused with flat editorial motion graphics. Use only deep red, scarlet, wine red, pure white, and black. Every character, prop, effect, title, and transition must remain a flat illustrated layer with clean line art, solid color fills, white nega

A retro American diner. A young woman sits at the counter, sipping a milkshake as a server approaches carrying coffee and a full breakfast. An elderly woman suddenly stands up and bumps into the server, sending coffee, eggs, bacon, and plates flying through the air. In that exact instant, the entire world freezes. Everyone and everything is completely suspen

Global setup: Create a 10-second vertical 9:16 high-fashion makeup editorial film. The adult fashion model from input[0] remains the same recognizable person in every shot: dark brown hair arranged in two neat crown braids, softly bleached brows, pale silver eye makeup, natural matte skin, and muted nude lips. Preserve her facial identity, hairstyle, makeup

Global setup: Create a 30-second vertical 9:16 first-person cooking-vlog comedy in a real home kitchen, compositing one flat 2D chibi sticker character into the live-action scene. The kitchen remains fixed: black iron wok on the stove, white ceramic tile wall, outlet, soy-sauce and cooking-oil bottles against the wall, stainless-steel sink on screen right, n

Global setup: Create a 30-second vertical 9:16 authentic smartphone travel vlog in Santorini, Greece. The female travel vlogger from input[0] appears in every scene with the same recognizable face, body proportions, clothing continuity, small tan crossbody bag, carefree personality, and observant reactions. Preserve her identity and the phone-camera look acr

Global setup: Create a 30-second vertical 9:16 Korean-language webtoon motion-comic trailer based on a fashionable aristocratic revenge drama. Use clean digital linework, elegant panel reveals, controlled gradients, selective animation, sharp reaction close-ups, and jewel-toned ballroom lighting. The young noblewoman from input[0] plays Deborah Seymour and c

Global setup: Create a 30-second 16:9 live-action neo-noir road-film sequence at dusk in heavy rain. The grand touring car from input[0] drives along one winding mountain road; preserve its recognizable paint color, trim, headlight shape, body geometry, wheel design, and plate layout in every shot. Dense pine forest rises on the inside of the bends; the outs

Please create a satisfying pastry-cutting ASMR video. At the beginning, the camera should be far enough away to clearly show the entire green cake, the full-textured white ceramic plate, and the simple light beige background. Do not start too close to the dessert. The composition should feel clean, balanced, and minimal, with the green cake centered on the p

Create a hyper-realistic ASMR video of a futuristic blue water-droplet mechanical keyboard. The keycaps are not normal round buttons — they should look and behave like translucent liquid water droplets, crystal-clear blue and cyan, soft, glossy, and slightly elastic, sitting on a dark reflective keyboard base with subtle blue backlighting. Show close-up mac

A soothing ASMR close-up video of a refreshing green iced drink in a clear glass on a clean white background. A small transparent glass spoon gently stirs the bright green liquid in slow circular motions. Large, clear ice cubes softly clink against the glass and spoon, while fresh mint leaves slowly move on the surface. Tiny condensation droplets form on the

A close-up ASMR video showcasing the preparation and presentation of innovative turkey dishes, including turkey noodles, turkey rice, and turkey skewers. The camera pans smoothly over a vibrant kitchen counter as hands skillfully assemble Thai-style basil turkey with fragrant herbs and spices, followed by a slow-motion shot of rice cake cheese turkey rice be

A large kitchen knife slowly slices through a thick rectangular bar of rainbow-layered soap. Each colorful layer separates cleanly as thin slices fall onto the cutting board. Capture smooth blade scraping, soft waxy cutting, tiny soap flakes, and satisfying muted slicing sounds. Bright clean studio lighting, macro shot.

Extreme close-up of two hands slowly pressing a sheet of clear bubble wrap on a clean white surface. The fingers press each bubble one by one in a neat sequence, creating crisp popping sounds. Some bubbles are pressed slowly, others more quickly for rhythmic variation. Capture sharp pops, light plastic crinkling, fingertip taps, and subtle hand movements. Br

Two hands slowly stretch, fold, press, and poke a thick glossy rainbow slime on a smooth tabletop. Large transparent air bubbles form and pop inside the slime. Capture sticky stretching, wet squishing, bubble pops, finger pokes, and soft suction sounds. Vibrant colors, extreme macro detail, controlled movements.

Extreme macro close-up of the side of a person's head during a careful ear-cleaning ASMR session. A gloved hand slowly guides a soft cotton swab visibly through the ear canal opening and gently rotates it inside the shallow outer ear canal, rather than wiping the surface of the ear. The camera angle clearly shows the cotton tip entering and cleaning the inne

Create a 60-second Japanese life-encouragement audio scene spoken by a middle-aged man. Voice direction: low, warm, steady, slightly weathered, with a kind uncle-like presence. The tone should be sincere and grounded, not motivational-speaker style. Scene: a quiet late-night neighborhood bar or small study after work. Add a low room tone, a faint cup sound

Create a 60-second English midnight radio companion audio scene with one female host only. Voice: adult female, warm low radio voice, calm, intimate, intelligent, slightly smoky, comforting. Scene: a quiet midnight radio studio. Add subtle radio texture, low room tone, light paper movement, and a peaceful late-night atmosphere. Keep background sounds gentl

Create a 60-second Japanese comfort audio scene from a young woman welcoming the listener home. Voice direction: bright, gentle, close, slightly smiling, with a soft affectionate tone. It should feel like someone waiting at the door after a long day, not childish or exaggerated. Scene: a small apartment entrance at dusk. The listener has just come home exh

Create a 40- to 60-second emotional Korean phone-call audio scene. Sound design: start with about two seconds of incoming phone beeps or a soft dial tone before the voice begins. After the call connects, apply a subtle realistic phone-call texture, not too distorted. Voice direction: one clean young male voice, intimate and sincere. He starts controlled an

Create a 60-second English female ASMR goodnight audio scene. Voice direction: one warm adult female voice, close to the microphone, soft, intimate, natural, and slightly sleepy near the end. Keep the performance beautiful and restrained, not theatrical. Scene: late at night in a cozy bedroom while gentle rain taps on the window. Add subtle blanket rustlin

Create a 60-second Japanese comfort audio scene spoken by one young woman on a late-night train. Voice direction: a young Japanese woman with a soft, clear, close voice. She sounds gentle, slightly tired, and quietly affectionate, as if sitting beside the listener on the last train home. Scene: a late-night train car with only a few passengers. City lights

Create a 60-second Japanese goodnight comfort audio scene spoken by one onsen inn owner. Voice direction: a warm middle-aged Japanese woman, elegant, gentle, attentive, and quietly nurturing. Her voice should feel beautiful, calm, and lived-in, like a thoughtful inn owner checking on a guest before sleep. Scene: late night at a traditional hot spring inn.

Create a 55- to 70-second English apology audio scene spoken by one young man who once bullied the listener. Voice direction: young adult male, fragile and broken, low and slightly rough, emotionally worn down. He sounds ashamed, hesitant, and painfully sincere, as if he has rehearsed this apology for years but still struggles to say it. Add small pauses, r

Please generate one continuous video strictly based on the uploaded 9-grid storyboard image input[0]. The 9-grid storyboard should be read from left to right, top to bottom, corresponding to Shot 1 through Shot 9. Each grid cell is not an independent image, but a continuous shot node within the video. Please connect all 9 shots naturally into one complete na

Please generate one continuous video strictly based on the uploaded **9-grid storyboard image** `input[0]`. The 9-grid storyboard should be read from left to right, top to bottom, corresponding to **Shot 1 through Shot 9**. Each grid cell is not an independent image, but a continuous shot node within the video. Please connect all 9 shots naturally into one

Create a clean 3x3, 9-panel anime storyboard contact sheet for a thunder iaido sword-draw duel.  Use input[0] as the visual reference for the protagonist. Keep the protagonist's face, hairstyle, age impression, body type, and overall identity consistent with input[0] across all panels.  Style: Japanese anime storyboard line art, simple clean contours, spar

Use input[0] as the protagonist visual reference. Preserve the referenced woman's identity-defining visual traits across all panels: face shape, hairstyle, age range, outfit vibe, body silhouette, and overall screen presence. Do not copy the input image background or pose literally; adapt only the protagonist identity into the storyboard. Create a single ci

Please generate a horizontal 16:9 overseas film-style storyboard image on the theme of "Deep-Sea Divers Encountering a Giant Sea Worm in an Ancient Underwater Civilization Ruin." The layout is a 3x3 nine-grid storyboard sheet with 9 panels total. Above each panel is the cinematic image; below it is a brief English description. Do not generate character desig

Create a 9-panel storyboard contact sheet for a thunder iaido sword-draw duel sequence. Style: clean Japanese anime storyboard line drawing, simple elegant linework, fewer strokes, clear contours, sparse shading, limited blue-gray color wash. Use electric blue lightning accents only where needed. Clean readable anime production storyboard feeling, polished

Please generate a horizontal 16:9 overseas film pitch deck storyboard image on the theme of "Abyssal Worm Encounter." The layout is a 3x3 nine-grid storyboard sequence page with 9 panels total. Maintain the formal, high-order sensibility of a professional film production board, but do not include complex A/B/C/D/E zone divisions, character turnarounds, route

Please generate a horizontal 16:9 overseas film-style storyboard image on the theme of "Being Chased by a Giant Sea Worm in Deep-Sea Ruins." The layout is a 3x3 nine-grid storyboard sheet with 9 panels total. Although the original reference may have been a horizontal 8-panel strip layout, this version must be converted to a nine-grid format. Do not generate

Use input[0] as the first frame. The woman on the rainy platform slowly turns her attention toward the distant train lights, then settles into a still, attentive pose. Keep the original wide framing and a locked camera. Let rain fall steadily beyond the canopy and warm reflections shimmer on the wet platform. Preserve the station architecture and the charact

Use input[0] as the visual reference for the character and rooftop. A medium shot at sunset: the character turns slowly from looking over the skyline to looking just past the camera, then holds a thoughtful gaze. A light breeze moves the jacket hem and a few strands of hair. Keep the camera fixed, the horizon level, and the sunset lighting consistent. Keep t

Use input[0] as the visual reference for the brass lantern and workshop. Reframe as a close-up of the lantern alone resting on the wooden bench; no person or hands in the shot. Move the camera slowly forward along a straight path, keeping the entire lantern inside the frame. The lantern remains stationary, its base touching the bench. Preserve its handle, gl

Use the woman in input[0] as the character reference. Reframe to a close-up of her head and shoulders alone in the evening cafe; leave the conversation partner outside the frame, just left of the camera. She listens with her mouth closed, briefly lowers her eyes, then looks back toward the same eye line with a small, relieved smile. Keep the camera locked an