やめ太郎「あー、パクられた！クソッ！」

プロンプト

Wan 3.0 実送信プロンプト（音声後編集は添付manifest参照） · revision 1

Create exactly one 24-second 16:9 Japanese office comedy video with synchronized Japanese audio. Polished cinematic 3D staging, preserving each reference character's own proportions and materials. Bright ordinary office, then an abrupt cut to a night street. No title card, subtitles, news graphics, background music or laugh track.

References:
Image 1: Yametaro, exact face, black hair, round black glasses with white lens areas, compact toy body, purple patterned shirt and black trousers.
Image 2: Takosan, exact toy-diorama character, matte black hooded robe, white face, solid black round eyes, two human-shaped arms ending in small round white fingerless hands. Exactly six slender lower-body tentacles radiate at 60-degree intervals, not six arranged in a front row. Preserve the sheet's small speaking mouth.
Image 3: Sobaya, exact muscular broad human body, short black hair, neutral gray exposed skin, white mask with its black and red markings, fitted white T-shirt, black trousers and white sneakers. Preserve broad pectoral shading and natural fabric folds, but no nipple protrusions, outlines, dots or localized nipple shadows through the shirt.
Audio 1: Yametaro voice-timbre reference only. Use this same voice for his visible first line and the final off-screen line. Do not copy reference words or timing.
Audio 2: Takosan canonical voice-timbre reference only, adopted from the finished episode 84. Match its vocal identity; do not copy its words, broken grammar or pauses. Deliver the new office line in fluent natural Japanese.
Use sheets only for character identity and clothing, not panels, labels, multiple copies, studio backgrounds or pose layouts.

Yametaro identity lock — Image 1 is the exact design authority, not loose inspiration. Preserve its head width-to-height ratio, hair silhouette and hairline, cheek and ear placement, short neck and compact body proportions in every frame. Keep the round black spectacle frames and the sheet's simple white lens/eye areas; do not invent visible pupils, irises, realistic eyeballs, eye sockets, eyelids or eyebrows. Keep the tiny black forehead mark separate from the low, subtle skin-colored nose; no black nose dot, outlined nostrils or projecting nose. Preserve the round cheeks and flat pink blush marks, smooth skin-colored chin, ears and nape, with no black contour bands or facial creases. Do not lengthen the forehead/head, inflate the hair, enlarge ears or add realistic lips. Animate the small inset mouth using the sheet's rest and A/I/U/E/O views as articulation guidance, with visible mouth interior when speaking; do not freeze the smile, protrude the lips or stretch the mouth into a new facial design. Emotion and cinematic lighting must not redesign the face. Wardrobe follows the scene instructions independently of identity. A previous generated take is not an identity reference.

Overall continuity: exactly one Yametaro and one Takosan seated side by side at adjacent desks in the office; Yametaro left, Takosan right. Each has one monitor. Yametaro scrolls a recognizable dark X social-media feed, with X mark, avatar rows and social icons. Takosan watches a cat video in a recognizable YouTube layout, red play mark, large cat player and thumbnail sidebar. Both are obviously slacking at work. Neither screen shows a spreadsheet, code editor or work document. No arrest imagery or arrest headlines before the final street cut. The cat video is muted. Maintain monitor ownership and positions. Sobaya appears only in the street scene; the office pair never appear in the street.

Natural speech takes priority over approximate shot timestamps. Shot time ranges are editorial guides, not speech-duration targets or mandatory utterance start/end times. Deliver every line at a natural conversational pace appropriate to its emotion. Do not stretch syllables, vowels, final words or pauses to fill a time window, and do not rush speech to hit a timestamp. Let each utterance finish naturally, then close the mouth. Use remaining time for silent reactions, glances, small gestures, ambience or the final action. Speak only the dialogue written in this prompt; do not invent additional lines during generation. Adjust internal cut boundaries to complete natural utterances while preserving scene order and the total duration of 24 seconds. The one intentional action cue is that Sobaya is fully restrained on the final phrase パクられたんや！

Shot 1 [0-2 s] — Medium-wide rear three-quarter view of both seated characters and both readable monitor layouts. Start directly in the office, no fade-in. Yametaro scrolls X; Takosan watches the cat. Yametaro suddenly stops moving his mouse. No speech yet.

Shot 2 [2-6 s] — Cut to a stable front three-quarter chest-up of Yametaro with his speaking mouth clearly visible. Irritated exclamation, an ordinary annoyed colleague rather than a screaming monster. Yametaro says exactly once: 「あー、パクられた！ クソッ！」. Make the shout forceful through voice, posture and reference-based mouth shapes; keep white lenses, head proportions and hair silhouette unchanged. Remain on him until the final phoneme and mouth closure.

Shot 3 [6-14 s] — Cut to Takosan chest-up, turning slightly toward Yametaro, with his small mouth clearly readable. His monitor continues showing YouTube. Takosan speaks naturally and with sincere mild concern, the exact Audio 2 canonical vocal identity, distinct from Yametaro, standard Japanese, understated and conversational. He says exactly once: 「どうしたんですか？ やめ太郎さんのアイディアが誰かにパクられたんですか？」. Preserve the toy face and animate only the existing small mouth without adding human lips or teeth. Finish naturally, close mouth, brief pause. Yametaro does not start his reply on this office image.

Shot 4 [14-24 s] — Hard cut BEFORE the first word of Yametaro's reply to a full-frame night street outside an izakaya, not a monitor inset. A low, mildly handheld medium-wide view clearly shows Sobaya's entire human body on all fours, supported on both palms and both shoe soles, knees lifted. A single overturned beer mug is already on the ground. He is drunk and violently animal-like in MOVEMENT ONLY: a low rapid quadrupedal charge, powerful alternating hand-and-foot contacts, abrupt sideways pivot, torso twisting away from approaching police, knocking over one lightweight chair. Keep his human anatomy, clothes, gray skin and white mask; no animal transformation, extra limbs, claws, muzzle, fur or tail. Three uniformed adult police officers close in, two from the sides and a third behind. Sobaya makes one strong twisting attempt to break their grip while staying low, then the officers collectively restrain his arms and torso against the pavement. Show clear separate limbs and credible contact; no interpenetration. His mask stays correctly attached throughout the charge, turn, struggle and restraint, never displaced, lifted, transparent or removed.
Over this ENTIRE street scene, Yametaro continues from the office as a clean off-screen voice, exactly once: 「いや、そば屋さんが酔って暴れて警察にパクられたんや！」. This voice belongs exclusively to Yametaro, never Sobaya or a police officer. At 「酔って暴れて」 show the quadrupedal rampage; at 「警察にパクられたんや！」 complete the restraint. Hold the restrained Sobaya for one brief silent beat while he still wriggles slightly and a parked police car's red lights rotate behind him. End here without returning to the office.

Audio: Yametaro has the Audio 1 voice, relaxed low-mid masculine voice with Kansai intonation, annoyed in the first line and matter-of-fact in the punchline. Takosan uses Audio 2 canonical voice, with calm conversational delivery. Pronunciation: やめ太郎=やめたろう; そば屋=そばや; 酔って=よって; 警察=けいさつ. Generate visible speech and mouth motion together for Shots 2 and 3, with closure during silence and no cut before line completion. The final Yametaro line is strictly off-screen voice-over, continuous and intelligible, not lip-synced to Sobaya. No intelligible police dialogue, no Sobaya dialogue. Office mouse clicks and room tone; street palm impacts, shoe scrapes, chair clatter, clothing struggle and distant siren mixed below dialogue. No music, no extra commentary.

Hard constraints: one continuous 24-second generated video with all four shots; original three dialogue lines exactly once, in order, no invented lines; only Yametaro and Takosan in office, only one Sobaya and three officers in street. Sobaya remains human and masked in EVERY street frame, performs quadrupedal locomotion rather than upright running or slow knee-crawling. His hands are free of props for locomotion. Preserve character identities. No early punchline reveal, no subtitles, no news overlay, no watermark, no duplicate characters, no sheet labels. X/YouTube interface cues are intentional in-screen graphics; do not add unrelated text elsewhere.