A tiny desk meme prompt describes the cramped office desk, five performers in studio headphones and the shot order: band shot, laughing close-up, sax reveal. The prompts below are a starting point for image and image-to-video models. Expect retries, because keeping five people recognisable across three shots is the hard part.
One clear adult per photo, 1 to 5 photos. Empty roles are played by the people you add. JPG, PNG or WebP, up to 10 MB; at least 300 × 300 pixels.
Add one to five photos, check who plays each role and create your vertical MP4.
Original sound from the reference clip is included by default. Choose Silent for a video without its soundtrack.
10 credits · Seedance 2.5 · 14s · vertical MP4 · See pricing
Attach your portraits in the order each prompt names. Describe the scene rather than naming real performers.
Photoreal still from a live office concert. A small band squeezed behind a cluttered wooden desk in front of floor-to-ceiling bookshelves packed with records, books and toys. In front, a singer seated at a red keyboard with a microphone on a stand; standing behind on the left and on the right, two backing singers. Everyone wears over-ear studio headphones with boom mics. The three performers are the people in the attached photos, in order, one person per photo, each in the clothes from their own photo. Soft daylight, vertical 9:16, every face sharp, no text, no logos.
Attach the lead first, then the left and right backing singers.
Photoreal close-up from the same office concert: one singer in over-ear studio headphones laughing with a huge grin mid-song, bookshelves softly blurred behind. The singer is the person in the attached photo, in the clothes from their photo. Head and shoulders, soft daylight, vertical 9:16, no text.
Photoreal close-up from the same office concert: a saxophone player in studio headphones playing a gold saxophone in front of shelves of records, with the seated singer soft and out of focus in the lower right corner. The sax player is the person in the attached photo, in the clothes from their photo. Vertical 9:16, soft daylight, no text.
Three-shot live office concert clip, 14 seconds, vertical 9:16. Shot 1, 0 to 5 seconds: the band from image 1 performs behind the desk, the seated singer sways at the microphone. Shot 2, 5 to 11 seconds: cut to the singer from image 2 cracking up laughing mid-song. Shot 3, 11 to 14 seconds: cut to the sax player from image 3 starting a solo. Keep every face, outfit and the room exactly as in the images, natural motion, slight handheld camera, no text.
Use the three stills above as images 1, 2 and 3.
Write five short captions for a tiny desk meme video starring [team or group]. The joke: [what happened]. The sax player is [name]. Keep each caption under 12 words, no hashtags.
Models fill gaps with generic studios, so spell out the room: a cluttered desk, bookshelves of records and toys, a red keyboard, headphones with boom mics and soft daylight. Then name the shot order, because the meme is a sequence and the sax has to come last.
Keep it vertical, ask for sharp faces and no text, and describe the scene instead of naming real performers. Prompts that name celebrities are often refused and are not needed for the joke.
The usual failures are the same person appearing twice, outfits drifting between people and a face that stops resembling its photo by the last shot. Generating one shot at a time, with one to three people per image, avoids most of them.
Sharp, front-facing portraits in daylight matter more than prompt wording. Sunglasses, heavy filters and group photos are where likeness slips first.
Make the three stills, pick the best of each, then hand them to an image-to-video model with the motion prompt. Budget several renders: the laugh often comes too early or the cut to the sax lands late.
Tiny Desk MeMe runs a tuned recipe for this exact clip. Upload one to five portraits, one per role, and you get the 14-second vertical video with the band shot, the laugh and the sax reveal, $9.99 for a 480p clip with the Single pack.
Start with what the meme means, then cast your band or compare the ways to make one.
Upload up to five portraits and get your band’s version of the Outstanding parody.
Up to five photos, five roles, sound and posting, step by step.
CapCut, editors, prompting a model yourself or one click.
What a chat can do, where a still falls short, and how to get the clip.
Twenty short answers about the meme, photos, sound and price.
Up to five photos in, a 14-second parody out, with the laugh and the sax reveal in place.
