Video trend
This trend is a chain: one phone moves through a packed room and eight people take the song in turn. Each of those eight comes from a photo you upload, so the whole group can be your friends, your team or your family. This page is about casting them: what each photo changes, what stays fixed and how to choose pictures that survive a moving camera.
Every role you see up close is replaced, so the studio asks for all eight and will not start without them. There are seven singers and one closer. If you have fewer people, you can use a photo twice, but that person will then appear twice in the same room, sometimes in the same frame.
Each person's face, hair, build and outfit come from their own photo. There is no costume, so nobody has to dress up, and the outfits in your video will be the ones in your pictures. A mix of everyday clothes makes people easier to tell apart as the camera moves.
One person in each photo, with a sharp face turned near the camera and the whole head and hair in frame. No sunglasses or caps, unless that is how the person always looks. Waist-up or full-body shots give the model more to go on, since most of the singers are seen from the waist up or wider.
Put your most committed performer first, because the opening seconds decide whether people keep watching. Save the closer slot for the person nobody expects to join in: a parent, a boss or the quiet friend. The closer is on screen for roughly the last ten seconds, longer than anyone else.
The white-walled room, the wide windows, the phone's path through the crowd and the order of the turns. Your video also keeps the original audio. People in the background are not cast, so they are kept from the reference or rendered by the model as extras.
You need a yes from all eight people. Use only photos of yourself or of people who agreed to be in the video, never a celebrity, an athlete or anyone who has not consented. Show them the result before you post it, and label it as AI-made.
Add a photo for each role shown below. All roles need a photo.
Add one clear photo per person, only of people who agreed to appear. A sharp face near the camera with the whole head and hair works best. Put the person nobody expects in the closer slot.
Choose 480p, 720p or 1080p and review the quote. It covers all eight photo preparations and all 37 seconds, so read it before you generate.
Pause on each of the eight turns, then check the joins at about 11.5 and 23 seconds, the hands near the lens and the mouths on the chorus before downloading your portrait MP4.
All eight people you see up close come from your photos. Each person's face, hair, build and outfit come from their own picture, so nobody needs a costume.
About 37 seconds in portrait, with one phone moving from person to person and no cuts to speak of. The room, the camera's path and the order stay the same.
Your video keeps the clip's own audio, including "Baby" by Justin Bieber featuring Ludacris. Nobody records anything, and the mouths follow the reference performance.
The closer stands beside the first singer from the first frame and takes the camera at the end. The preview is the reference clip, not an Oakgen result, and its eight main performers are replaced by your photos.
It is a 37-second portrait video in one long phone take. Seven people lip-sync the same pop song in turn in a crowded room, then the camera lands on an eighth person while everyone laughs. Oakgen's template replaces all eight with people from your photos.
"Baby" by Justin Bieber featuring Ludacris, identified from its lyrics. Neither artist appears in the clip or is connected to this template. We did not find an established name for this exact clip, so Oakgen calls it Baby Lip Sync.
Yes. The source cast is never kept, so the studio will not start until all eight slots have a photo. You can use a photo twice, but that person will appear twice in the same room.
The person nobody can picture doing this: a parent, a boss or the quiet friend. The closer stands beside the first singer from the start and takes the camera for about the last ten seconds.
No. The video keeps the clip's original audio and nobody records anything. The mouths follow the reference performance, so check them closely before posting.
Yes. There is no costume. Each person's outfit comes from their own photo. The template asks for no logos or lettering from the source clothes to carry over.
They are extras and are not cast from your photos. The template asks for anonymous people in the background, but it is an instruction to a model, so check the background before you post.
Only photos of yourself and of people who agreed to be in the video, all eight of them. Do not upload photos of athletes, celebrities or anyone who has not consented, and label AI videos clearly when you share them.
A single model request cannot cover 37 seconds, so the template makes three parts and joins them near 11.5 and 23 seconds while the camera is swinging between people. Check those moments in your video.
The template is written for people lip-syncing. An animal photo is not blocked, but the result is untested and the mouth timing may not look right.
About 37 seconds in portrait 9:16.
No. It is the reference clip the trend comes from, not an Oakgen result. Review every face in your video before sharing.
5,231 credits per 720p video. The quote includes all eight photo preparations and the complete video. Check it before generating.
Plan a story, keep your character recognizable, and make the finished clip your own.
Make the Baby Lip Sync AI trend step by step: choose your eight people and photos, assign the slots, read the quote, check all the faces and post your group video.
What the Baby Lip Sync AI trend is, why a 2010 pop song works in a crowded room, why the last person makes the video and how the template rebuilds it from 8 photos.
Upload 8 photos. Everything else is set.
Oakgen.ai