You provide
- Photo 1: the person on the keys
- Photo 2: the hype partner
- Nothing else: no prompt, no green screen and no editing
AI video generator · 2 photos · 1 duet
DanseAI is an Alors on Danse AI video generator: upload two photos of people and it recreates the white-studio keyboard meme with your faces. Photo 1 plays the mini keyboard on the left; Photo 2 is the hype partner who gets the close-up. 5 to 30 seconds, 9:16 or 16:9, up to 1080p, soundtrack included, no prompt needed.
Make my Alors on Danse video

10 seconds · 9:16 · 720p · soundtrack included
Faces and outfits can drift between cuts, especially in the close-up. Watch the whole clip before you post it.
Your two photos replace the performers in this reference. Play it with sound to hear the soundtrack added to your video.
Everything the generator does, one fact per row. Prices are the defaults; the generator always shows the exact total before you confirm.
Photo 1 replaces the performer on the left, who stays at the table and plays the mini keyboard and laptop. Photo 2 replaces the performer on the right, who points, throws an arm up, coaches the beat and then leans right into the camera for the close-up. Other Alors on Danse tools number their photos differently, so check the labels here before uploading, or use Swap photos to switch roles.


Put the more expressive face in Photo 2: that role gets the big close-up, so a sharp, front-facing photo matters most there. Waist-up photos give the model your outfit; a tight face crop leaves clothes to chance.
JPG, PNG or WebP, up to 20 MB each, with the shorter edge at least 300 px. One person per photo; crop out anyone else.
Only upload photos of people who agreed to appear in the video, and label the result as AI-generated when you post it.
Add Photo 1 for the keyboard player and Photo 2 for the hype partner, one person per photo.
Pick 5, 10, 15 or 30 seconds, a Seedance or MiniMax H3 model and 480p, 720p or 1080p where the model offers it. The credit cost updates before you generate.
Most videos take a few minutes and you can leave the page. Check both faces through the close-up, then download the MP4 with its soundtrack from your history.
Seedance 2.0 Mini is the default because it costs the fewest credits. Seedance 2.5 gives the best likeness: it checks both photos and the reference clip first, then renders even 30 seconds in one pass. MiniMax H3 balances likeness and price; Seedance 2.0 Fast and 2.0 sit in between. Mini stops at 15 seconds; for 30 seconds the others join two 15-second generations, so faces can shift slightly at the join. Mini and Fast stop at 720p, and MiniMax H3 offers 720p and 1080p. No model guarantees a perfect likeness, so try a short clip first if you are unsure.
Default and cheapest; stops at 15 seconds; holds likeness less well
Lower price at 30 seconds; up to 720p
Up to 1080p
Best likeness; checks both photos and the reference first
Balanced likeness and price; 720p and 1080p only
Alors on Danse videos use site credits. The total depends on the length, model and resolution, is shown before you generate and is charged once per finished video.
| Model | 10 seconds · 480p | 10 seconds · 720p | 10 seconds · 1080p |
|---|---|---|---|
| Seedance 2.0 Mini | Costs 11 credits | Costs 24 credits | — |
| Seedance 2.0 Fast | Costs 40 credits | Costs 87 credits | — |
| Seedance 2.0 | Costs 68 credits | Costs 145 credits | Costs 364 credits |
| Seedance 2.5 | Costs 97 credits | Costs 219 credits | Costs 388 credits |
| MiniMax H3 | — | Costs 97 credits | Costs 154 credits |
A 5-second video starts at 6 credits. Check the total for your length, model and resolution before generating. Eligible new accounts currently receive 3 credits, usable on Seedance 2.0 Mini; other models need purchased credits. Signup credits expire after 7 days.
Free accounts receive a moving watermark. Any purchase unlocks the saved clean original, including videos you have already made.
If the generation task fails or a photo is not accepted, the credits charged for it are returned. A finished video can still contain visual imperfections, so check it before sharing.
Alors on Danse videos use site credits. The total depends on the length, model and resolution, is shown before you generate and is charged once per finished video.
Starter
For trying the trend
240 credits / month
≈ 10 ten-second Seedance 2.0 Mini videos
Pro
PopularFor regular creators
780 credits / month
≈ 32 ten-second Seedance 2.0 Mini videos
Enterprise
For agencies and pages
2,650 credits / month
≈ 110 ten-second Seedance 2.0 Mini videos
Uploading photos and seeing the price are free. Generating uses credits, from 6 credits for a 5-second 480p video. Accounts that have never paid download the full-length MP4 with a moving watermark; any purchase, including the $5 single-creation trial, removes it from every video in your history. Failed videos are refunded.
Free vs paid, line by lineYes. The model renders silent video, and the finishing step adds the reference clip’s own soundtrack, cut to the length you chose, so the MP4 you download has sound. The music is Stromae’s commercial recording: platforms may treat your post like any other clip with commercial music, and some creators add the platform’s official sound when posting instead.
How to post it on TikTok or ReelsThe template is a vertical Kratos and Zeus version of the white-studio scene. Your two photos replace both performers. The 5, 10 and 15-second options are the opening of the same clip; 30 seconds adds the ending where both walk off across the studio, holding the last frame for about a second. The model receives this same clip as its motion and camera reference.
The format goes back to a comedy sketch from Jamel Debbouze’s Made in Jamel, in which Jamel walks into Stromae’s white studio and pretends to coach him into writing “Alors on danse”. It is often captioned as the song’s making-of, but Stromae wrote the song on his own; the sketch is a joke. AI remakes now swap new pairs into the two roles, from tech CEOs to film characters, which made the Alors on Danse AI trend take off on TikTok. It works because the calm one at the keyboard and the over-committed hype partner are easy to recast with friends, partners or family.
The title is French for “So we dance”. Stromae, who is Belgian, released the song in 2009.
Who is the other guy? The meme explainedWhat “Alors on danse” means
This is an independent video tool, with no affiliation to Stromae, Jamel Debbouze, their labels or the creator of the reference remake.
The same white-studio scene with different pairs of characters. Characters and likenesses belong to their respective owners.
The Lord of the Rings: MC Gandalf x DJ Frodo
Harry Potter. MC Dumbledore x DJ Harry
Star Wars: MC Palpatine x DJ Anakin
Spider-Man. DJ Tom x MC Tobey x Andrew
3 people: this tool takes two photos
DJ Yamal x MC Messi x MC Ronaldo
3 people: this tool takes two photos
Breaking Bad
Behind the Scenes "Alors on danse"
Photo 1 always takes the keyboard. Use Swap photos and generate again.
Use a sharper, front-facing photo with even light and no sunglasses or mask. Seedance 2.5 and 2.0 usually hold identities better than Mini.
If a photo is not accepted or the model fails, the credits for that video are returned automatically. Try a different everyday photo.
Every finished MP4 includes the soundtrack. Check that the player is not muted and that the iPhone silent switch is off; if a platform strips the audio, add its official sound.
On every model except Seedance 2.5, 30 seconds joins two 15-second parts. Use Seedance 2.5 to render 30 seconds in one pass.
Photo 1 replaces the performer on the left who plays the mini keyboard. Photo 2 replaces the performer on the right who hypes the room and leans into the camera for the close-up. Use Swap photos to switch them.
No. The studio, moves, camera and role instructions come with the template. Upload two photos, choose the settings and generate.
Yes. The finished MP4 includes the reference clip’s soundtrack, cut to your chosen length. It does not generate a new song or clone a voice. The music is a commercial recording, so platforms may handle it like other clips with commercial music.
Choose 5, 10, 15 or 30 seconds in vertical 9:16 (the default) or 16:9. Seedance models offer 480p and 720p; Seedance 2.0, 2.5 and MiniMax H3 also offer 1080p, and MiniMax H3 has no 480p. Vertical videos are 480 × 854, 720 × 1280 or 1080 × 1920; 16:9 videos are 854 × 480, 1280 × 720 or 1920 × 1080. The reference is vertical, so 16:9 widens the studio and can follow the moves less exactly. The default is Seedance 2.0 Mini, 10 seconds, vertical at 720p.
A 5-second video starts at 6 credits. Check the total for your length, model and resolution before generating. Eligible new accounts currently receive 3 credits, usable on Seedance 2.0 Mini; other models need purchased credits. Signup credits expire after 7 days.
Free accounts receive a moving watermark. Any purchase unlocks the saved clean original, including videos you have already made. A 5-second video starts at 6 credits. Check the total for your length, model and resolution before generating. Eligible new accounts currently receive 3 credits, usable on Seedance 2.0 Mini; other models need purchased credits. Signup credits expire after 7 days.
Exact likeness is not guaranteed. Faces, hair and clothing can change, especially in the close-up and around cuts, and the two identities can occasionally blend. Use clear photos and watch both people through the whole clip.
Usually a few minutes, longer for 30 seconds or busy periods. You can leave the page; finished videos are saved in your history.
If the generation task fails or a photo is not accepted, the credits charged for it are returned. A finished video can still contain visual imperfections, so check it before sharing.
A CapCut template places your clips or photos into a fixed edit. This tool uses an AI video model to recreate the whole performance with your two people, so they move, play the keys and lean into the camera themselves. It is not affiliated with CapCut.
No. There is no live filter for this meme. A filter tracks one face in your camera; this video re-renders two whole performers, their moves and the close-up from two photos, which takes an AI video model a few minutes.
No. A face swap pastes a face onto existing footage. Here the model generates the whole performance again from your photos, including bodies, outfits and the camera move.
Yes. The default 9:16 MP4 is ready for TikTok, Reels and Shorts. The soundtrack is Stromae’s commercial recording, so platforms may handle it like other clips with commercial music; label the video as AI-generated and only use photos of people who agreed.
No. General video apps need you to supply the reference clip, describe both roles in a prompt and add the music yourself. DanseAI bundles the reference clip, the role mapping, the model choice and the soundtrack; you only upload two photos.
No. DanseAI is an independent tool with no affiliation to Stromae, Jamel Debbouze, their labels, the creator of the reference remake, TikTok or CapCut.
You need a free account to generate, so your videos are saved in your history. No card is needed to sign up; you only pay when you buy credits. It works in any mobile or desktop browser.
No. Your photos are used to make your video, stored under random, unlisted file names and never added to any public examples. Your history and downloads require your account.
Two photos, a few minutes, one Alors on Danse video your friends will replay.
Make my Alors on Danse video