All guides

Tiny Desk AI Video Template: One Fixed Set, Your Faces

Published 1 November 2026

If you came looking for a tiny desk AI video template to download and fill in, there is not one, and that turns out to be the point. The performance is already made. The five camera angles, the cuts between them, the music and the length are all settled and fixed. What you supply is faces.

So is there a template or not

It depends what you mean by one. There is no file to download, no project to open, no scene to build. There is one performance, already made, and it runs the same way every time. In the sense most people mean when they ask, yes: the hard part is done and you are only putting people into it.

What there is not is a choice of templates. There is one scene, one piece of music and one length, and that is the whole catalogue. If you were hoping to browse a shelf of performances and pick the one you liked, this is not that, and no setting anywhere on the site opens one up.

Watch one first

The clip below is a real result rather than a mock-up. The five faces in it started as fictional portraits, which is why we are able to show them next to the video they produced. Sound is on, so check your volume.

  • Lead singerLead singer
  • KeysKeys
  • SaxophoneSaxophone
  • Backing, leftBacking, left
  • Backing, rightBacking, right
A finished video and the five portraits that went into it.

What is already decided for you

  • Five musicians play behind a desk in a cramped office, to a small room.
  • The camera visits five angles, and it cuts between them at the same moments every time.
  • One piece of music runs underneath. The same one on every video, and it cannot be swapped out.
  • A video runs just over 19 seconds, at 24 frames a second.
  • It comes out vertical, shaped for a phone screen.

None of that is adjustable, before or after, and none of it is hidden behind a setting somewhere.

What you actually supply

Two things. The faces, and the size.

  • Up to five portraits, one for each seat in the band: lead singer, keys, saxophone, and backing singers left and right. Only the lead has to be filled.
  • The size, which is the one setting there is. The smaller one is clear enough on a phone. The larger one holds up better on a laptop screen or a television.

Seats you leave empty are played by fictional portraits the site keeps for that purpose, so one photo still produces a five-piece band. Filling more seats costs nothing extra.

Nothing to write and nothing to cut

The part people dread about making a video like this is the part that is missing here. You do not describe the scene in words. You do not choose a shot list, set a length, pick music, or line anything up in an editing app. Put the portraits in their labelled seats, choose a size, and wait.

That is also why there is very little to get wrong. The one mistake worth watching for is a photo in the wrong seat, because the person in the saxophone seat gets handed a saxophone and there is no changing it afterwards. Read the labels once more before you start it.

What the one set costs

Credits are bought once and there is no subscription: 10 for $12.99, 30 for $29.99, 60 for $49.99. A video spends 10 credits at the smaller size and 15 at the larger one. If one does not come out, the credits come back by themselves.

Two files per video: one with the music on it, one silent, for when you would rather put your own sound under it in an editing app. We looked for a watermark frame by frame in the videos we downloaded and did not find one.

Across the videos we timed, the smaller size came back in roughly four to six minutes. The larger size is a great deal slower, and we have a stopwatch figure for one run of it only, so treat it as a long wait rather than a number.

When a fixed set is the wrong answer

One pre-built performance takes every decision off your hands, which also means you do not get to make any. If any of the following matters to you, this is not the tool.

  • You want a particular song. The music is part of the fixture, so that is the one thing the site definitely cannot do for you.
  • You want it longer, shorter, or cut differently. The shots and their timing do not move.
  • You want another setting, a stage, a street, a living room. There is one room.
  • You need the faces to be exact. They come out resembling the photos rather than reproducing them, and the movement and mouth shapes are approximate too.
  • You are posting somewhere with its own rules about music, or putting a real person in the band. Those are two separate permissions and this site hands you neither.

If you want to try it

The band builder is on the front page and the credit packs are on the pricing page. Watch the sample at full volume first. If the length and the fixed camera work are not what you had in mind, no photo you upload will change them.

Make a videoSee the packs

Sign in to continue

No password. What you entered stays on the page

Get credits

One video

$12.99

$12.99 per video

10 credits, never expire

  • About 19 seconds · 9:16 vertical MP4 · 480p or 720p HD
  • All five roles included; extra bandmates never cost more
  • Failed generations return their credits

3 videos

$29.99

$10.00 per video

30 credits, never expire

  • About 19 seconds · 9:16 vertical MP4 · 480p or 720p HD
  • All five roles included; extra bandmates never cost more
  • Failed generations return their credits

6 videos

Lowest per video

$49.99

$8.33 per video

60 credits, never expire

  • About 19 seconds · 9:16 vertical MP4 · 480p or 720p HD
  • All five roles included; extra bandmates never cost more
  • Failed generations return their credits

Prices are in USD. Per-video amounts are rounded; applicable taxes are additional. One-time packs, no subscription. Refund policy

Music and likeness rights are separate. Read the Terms of service.