Can I add a sixth person?

No. The reference has five performers, and the seats map one-to-one onto them.

Can I choose only the saxophone player?

Yes. Every seat is independent, so a sax-only recast is a valid render.

Does adding people change the render time?

Yes. Five people means five shots instead of two, so it costs more and takes longer than a solo render.

Can two people share one seat across two videos?

Yes. Seats are per render, so you can make one version with each person.

Short answer

Tiny Desk AI has five optional seats: lead singer, keyboard, backing vocal left, backing vocal right and saxophone. Fill one and only that performer changes; fill all five and the whole band is replaced. Price moves in three tiers — 1 person, 2–3 people and 4–5 people — and depends on resolution.

Make my band video

1 to 5 people: how the seats work

A solo lead and a five-piece band are the same product, the same price and the same three steps.

Updated 2026-10-29

The clip is not one shot. It is five shots cut together, and each seat only appears in some of them. Filling a seat therefore does not just change a face — it decides which shots need to be rendered at all.

The five seats

1.Lead singer
Shots 1 and 3
2.Keyboard
Shots 2, 4 and 5
3.Backing vocal · left
Shots 2 and 5
4.Backing vocal · right
Shots 2 and 5
5.Saxophone
Shot 4

Seat	Appears in	Screen time
1. Lead singer	Shots 1 and 3	About 8s, including the close-up
2. Keyboard	Shots 2, 4 and 5	Wide shots plus the sax solo corner
3. Backing vocal, left	Shots 2 and 5	Wide shots only
4. Backing vocal, right	Shots 2 and 5	Wide shots only
5. Saxophone	Shot 4	About 3.4s solo

Three practical casts

Solo — one photo

Fill only the lead singer seat. Two shots are rendered instead of five, so it is the fastest and the safest first try. The rest of the band stays exactly as in the original, which reads as “same band, new singer”.

Trio — three photos

Lead, keyboard and saxophone. This cast touches four of the five shots and covers every close-up, so the result feels like a new band rather than a new singer.

Full band — five photos

All five seats. Every shot is re-rendered, the whole line-up changes, and this is the version people recognise as the meme. It takes the longest because five shots are generated.

For photo advice per seat, see Tiny Desk AI photo requirements.

Why a solo costs less than a full band

Credits are priced per finished video by cast size and resolution: 7 to 25 credits. A solo lead only needs two shots rendered, so it is the cheapest tier; a full band needs all five. That is why the price moves with people, and why a five-person band costs more than a solo lead.

One rule: one face per seat

Never put the same photo in two seats. When one identity is asked to play two roles in the same shot, the model tends to replace neither of them and still reports success. Five seats means five different people.