Prompts Tiny Desk AI e guia de cenas

IDENTITY FIRST: the lead singer in this video is the person in [Image 1] and nobody else. [Video 1] gives only the camera position and the timing; the man singing in it is a placeholder who never appears: not that man's face, hair, skin, jacket or any logo.

A viral office-desk soul concert, part 1 of 2: a five-piece band plays an upbeat funk-soul song crowded around a wooden desk, and the person in [Image 1] sings lead.

THE LEAD SINGER IS THE KEY FACE OF THIS PART: in every close-up (0-1.3 s and 6.7-13 s) the face filling the frame is the person in [Image 1], with the face, hair color, hair length and style, hairline, skin tone, age, beard or no beard and clothes of [Image 1], from the first frame of each shot to the last, in profile and three-quarter view, eyes closed, mouth wide open while singing and head tipped back while laughing. If [Image 1] shows no clear top, the lead wears a plain solid-colour T-shirt; never a zip jacket and never a logo. In [Video 1] those close-ups show a stand-in, a man with short brown hair and stubble in a navy zip jacket: replace that stand-in completely. No navy zip jacket on the lead, and no short brown hair or stubble unless [Image 1] shows them. Only the camera angle, the timing and the mouth shapes come from [Video 1]; the head on screen is always [Image 1]'s.

Vertical 9:16, photorealistic, like a real live session filmed in an office: slightly handheld camera, natural warm lamps plus soft daylight, gentle film texture, real skin detail.

WHO IS WHO: the LEAD SINGER, standing at the front corner of the desk with a thin headset microphone, filmed in close-ups is the person in [Image 1]. Each of them appears in every shot they are in. Each one keeps the exact face from their own image (face shape, eye shape and color, eyebrows, nose, mouth, skin tone, age, hair color, hairline, hair length and style, and any beard, glasses or freckles), from the first frame to the last, also with the mouth wide open while singing and while laughing, so their friends recognize them instantly.

The rest of the five-piece band are invented adult strangers with their own faces, never a face from the images: an invented adult saxophone player with a gold tenor saxophone at the right side; an invented adult keyboard player seated at the red stage keyboard on the desk; an invented adult backup singer standing behind the desk on the left; an invented adult backup singer standing behind the desk on the right.

MOTION REFERENCE: [Video 1] is the reference performance for this part and has the same length. Re-film it shot for shot: the same hard cuts at the same moments, the same camera angles, framing, lens and handheld sway, the same positions around the desk, and every gesture, nod, fist pump, laugh, mouth shape and breath at the same moment as in [Video 1]. The people in [Video 1] are stand-ins who never appear in the video. Replace each one completely, face, hair and clothes: the man with short brown hair and stubble in a navy zip jacket in the close-ups becomes the lead (exactly the face from [Image 1]); the man with dark curly hair and a beard in a black T-shirt at the red keyboard becomes the keyboard player; the woman with dark curly hair in a mustard-yellow top behind on the left becomes backup singer 1; the bald man with glasses in a red polo shirt behind on the right becomes backup singer 2. Never copy a face, hairstyle, hair color, skin tone, body shape or outfit from [Video 1]: faces and hair come only from the images and clothes follow the WARDROBE line. The room in [Video 1] only guides the layout; build the room described below, with no readable signs, logos or text.

THE PLACE: a cramped office corner turned into a tiny concert stage. Behind the band, worn wooden bookshelves run to the ceiling, packed with vinyl records, old books, small trophies, toy figures, mugs and potted plants. In front, a scuffed wooden desk holds a red stage keyboard, a few coffee mugs and boom microphones on thin black stands. Everyone wears over-ear studio headphones and sings into boom microphones on thin black stands.

WARDROBE: each person from an image wears the clothes from their image (any logo or print becomes plain); the invented bandmates wear casual everyday clothes.

MOOD: Joyful, loose and soulful, like the last song of a great live set.

SHOTS (13 s, overall 0-13 s of a 19 s video): 0-1.3 s: a tight profile close-up of the lead (exactly the face from [Image 1]), with the hair and clothes from [Image 1] from the side, headphones on, a thin headset mic at the cheek, the shelves soft behind. The intro is instrumental and the lead is not singing yet: lips relaxed, eyes lowered, nodding on every beat, one open hand lifting to shoulder height and dropping, counting in. 1.3-3.2 s: hard cut to a medium shot from the front: the keyboard player front and center behind the red keyboard, backup singer 1 behind on the left, backup singer 2 behind on the right, the packed shelves filling the top of the frame. The band shouts the two-word hook together, in time with the voices: two short hits, mouths wide open, eyes squeezed, the keyboard player pumping both fists at chest height on each hit, backup singer 2 punching one fist up. 3.2-6.1 s: same band shot, the voices stop and the groove plays on: mouths closed in big grins, the keyboard player laughs and plays a few chords with the right hand, nodding on every beat; the backup singers sway side to side on the beat with small fist bounces at the chest, trading smiles. 6.1-6.7 s: the groove drops to a short break: the band leans in, still and smiling, waiting for the lead. 6.7-8.3 s: hard cut to a front three-quarter close-up of the lead (exactly the face from [Image 1]), with the hair and clothes from [Image 1], face large and sharp, eyes looking up past the camera: the lead sings the short rising phrase, lips in sync with the voice, mouth opening wider and eyebrows lifting as the pitch climbs. 8.3-9.6 s: the voice stops and the lead bursts into a huge, silent laugh (no laughing sound, the music carries it): head tipped back, eyes squeezed shut, shoulders bouncing on the beat. 9.6-11.2 s: still laughing, the lead turns to face the camera with a wide toothy grin, then raises one open hand in front of the chest and flutters the fingers in a little wave on the beat. 11.2-13 s: the lead's voice comes back in: the next phrase is sung with the head tilted up and the mouth wide open, one fist at the chest, lips in sync, and on the held note the eyes close into a grin; the note is still ringing.

Keep the music driving through the final frame, mid-phrase, with no ending, fade or silence.

AUDIO: [Audio 1] is the exact soundtrack of this part, from its first second to its last; keep it as the video's sound and act to it. The song is about 99 beats per minute (a beat every 0.6 s): nods, sways, fist pumps and claps land on its beats. The voices in [Audio 1] belong to whoever the shot list says is singing at that moment, with lips, jaw and breath in sync with each syllable; when no voice is heard, nobody's lips move as if singing. The saxophone in [Audio 1] is played by the sax player, fingers in time with the notes. Do not add any other music, voice, laugh or dialogue. [Audio 1] is the same sound as in [Video 1], so the mouths and moves in [Video 1] already match it.

Exactly five performers, all adults. No famous people, no children. No readable letters anywhere: no sign, poster text, show name, logo or watermark, no captions. Plain clothes with no logos, team names or numbers. One head per person, five fingers on each hand, never merge bodies, never morph or swap faces.

FINAL CHECK: every close-up of the lead shows the face and hair of [Image 1], recognizable at a glance; a face, hairstyle or jacket that looks like the man in [Video 1] is wrong.