The crowd chanting your name in AI video

Your selfie goes into two Seedance 2.5 nodes that put you on the giant screen with your first name written under it, seen from the crowd. Workflow visible here, no account needed.

Interactive demo

Real workflow, real results. Pan and zoom freely.

Use this workflow

The real workflow: the selfie wired as a reference into two Seedance 2.5 nodes, one venue per node. Both videos are the demo's own, the name written by the model.

At a glance

What the workflow produces: 2 videos of 10s in 720p, 9:16
Canvas structure: 6 nodes wired by 4 connections
Models wired in: Seedance 2.5
Cost of one full run: about 240 credits, roughly $2.88

Your face on a stadium's giant screen, your first name written under it, and the crowd on its feet chanting. This is the gift use of the family: a birthday, a leaving do, a question asked in public.

The unknown of this workflow was the text. Writing a clean word inside an image is something our IMAGE models have done for a long time; our video models, that had not been measured. It has been since 13 September 2026: Seedance 2.5 writes a short first name in capitals, clean, on a screen.

So stay inside what was measured. A short first name, in capitals. A whole sentence, a long name or an accented script have not been checked, and a video model that fails at text does not half fail: it returns a mush of letters. Both shots cost two hundred and forty credits.

How it works

1
Give one sharp selfie
Framed from the waist up, facing the camera, in daylight. It is the shot that appears in close-up on the screen, so its sharpness shows at scale.
2
Swap the first name
Both prompts carry FRANK in capitals. Put the name of the person the video is for, short and in capitals, and nothing else.
3
Pick the venue
A stadium at night, a concert, a hall. The venue changes the emotion: the stadium says achievement, the concert says celebration.
4
Look at the text before you send it
It is the one element of the shot that can fail outright. A glance at the video before you hand it over avoids sending a badly formed name.

Variations and use cases

The same canvas, tuned for different needs: each variation is two or three lines of prompt away.

The birthday present

A whole stadium chanting someone's first name costs two hundred and forty credits and gets watched ten times over. It is the use this workflow exists for.

A colleague leaving

Replacing the phone-filmed goodbye with a giant screen showing their name gives a keepsake nobody forgets, and one that circulates around the team.

Welcoming a new hire

A company greeting a new recruit by name on a stadium screen says its enthusiasm better than a standard welcome message ever does.

The spectacular question

A proposal, an invitation, a question asked in public: the giant screen gives a short sentence a solemnity that plain text does not have.

A club celebration

An amateur club putting its players on a big stadium screen gives each of them thirty seconds of the professional treatment, for the price of a coffee.

The job rather than the crowd

If the situation should change rather than the recognition, your future job places you in the setting of the role instead of in front of a stand.

Common mistakes

Writing a whole sentence

What was measured is a short first name in capitals. A sentence leaves the verified ground, and that is where a video model returns deformed letters that ruin the whole shot.

Using a long or accented name

The longer the word, the more chances the model has to get it wrong. Accents were not checked either: if the name carries any, look at the render before you hand the video over.

Asking for several texts

A name and a date, or a name and a logo, double the risk. One word on screen is the prompt that was measured, and it is the one that comes out clean.

Not looking before sending

This is the one workflow of the family where an element can fail outright. Thirty seconds of checking avoids giving away a video with a badly written name.

Recommended models

Frequently asked questions

Does the model really write my name?

Yes, if it is short and in capitals: that is what was measured on 13 September 2026 on both demo shots. A long name, a sentence or accents fall outside the verified ground, and there the text can deform.

Can I write a whole message?

That is not what was tested, and it is the case where a video model fails most often. If you want a sentence, generate the shot without text and add it in the edit, where it will be sharp for certain.

Can I use someone else's face?

Yes, with their consent: that is exactly the gift use. The workflow changes nothing about image rights, and a video made for someone is made with their agreement if it is going to be published.

What do the two shots cost?

Two hundred and forty credits, one hundred and twenty per five second video in 720p with sound. Changing the name and rerunning one shot costs one hundred and twenty.

Does the name really come out legible?

On both demo shots, yes: FRANK in clean capitals on the stadium screen and on the LED wall. It is a measurement from 13 September 2026, and it holds for a short word in capitals.

Can I change the script of the name?

A Latin-alphabet name has been verified. Non-Latin scripts have not been tested on this model, and any doubt about the text is settled by looking at the render before sending it.

What is the output format?

Vertical 9:16 in 720p, the native format of social feeds. The format and resolution guide gives the expected sizes network by network.

Can I use someone close to me?

Yes, with their consent, and that is the intended use. The workflow changes nothing about image rights: a video made for someone is made with their agreement if it is to be published.

The crowd chanting your name in AI video

Use this workflow

Free account, no card required. The workflow opens pre-filled in your canvas.