Oakgen.ai
The Baby Lip Sync AI trend puts your whole group in one long phone take. The camera moves through a crowded room, seven people take turns lip-syncing, and then it lands on the eighth person, the one nobody saw coming. You supply eight photos and the video does the rest. This guide goes from picking the people to posting, in order. For what the clip is and why it works, read the Baby Lip Sync trend explained.
Get your whole group in the take
Add eight photos, review the quote and make the 37-second Baby Lip Sync video with the original audio.
What the template gives you
One portrait 9:16 video, about 37 seconds long at 24 fps, with the clip's original audio. You replace eight people. The room, the camera's path and the order of the turns are fixed.
| Time (about) | Who has the camera | Worth knowing |
|---|---|---|
| 0 to 5 s | Singer 1 | The closer stands right next to them |
| 5 to 8 s | Singer 2 | Short turn, big energy |
| 8 to 12 s | Singer 3 | Makes a heart with both hands |
| 12 to 16 s | Singer 4 | Steady medium shot |
| 16 to 20 s | Singer 5 | Wide frame, hand gestures |
| 20 to 24 s | Singer 6 | Closest to the lens, hands up |
| 24 to 27 s | Singer 7 | The last singing turn |
| 27 to 37 s | The closer | Then the whole group laughing |
The timings are rounded from the reference clip. The turn-by-turn breakdown has more detail.
Step 1: Choose your eight
Start with the closer, the person nobody can picture doing this. Everything builds to them. Then give the first slot to whoever will look most committed, since they decide whether people keep watching. Everyone has to agree, so ask before you start. Never use a celebrity, an athlete or anyone who has not consented.
Step 2: Collect the photos
Each photo should have one person in it, and what matters is, in order:
- A sharp face near the camera. Most singers are seen from the waist up, so the face stays small enough to need detail.
- The whole head and hair. Hair is part of what makes people recognisable when they turn.
- Their real outfit. There is no costume, so the clothes in the photo are the ones in the video.
Avoid group photos, sunglasses, caps and heavy filters. The casting guide goes into more detail on what each photo controls.
Step 3: Assign the slots and read the quote
Open the Baby Lip Sync template, add a photo to each slot and pick 480p, 720p or 1080p. 720p is preselected. The template will not start until all eight are in. The studio shows the price before you generate, and this is one of the bigger jobs in the gallery because it is long and has eight people, so read the quote. A low-quality run is the cheapest way to test your photos.
Step 4: Know what to expect
We have not timed a run, so we will not promise a wait. Three things to know before it lands:
- It is made in three parts. They join around 11.5 and 23 seconds, while the camera is swinging between people.
- The background is not cast. Other people in the room are extras, and they are not your friends.
- The preview is the reference. The trend page shows the clip the trend comes from. We have not reviewed a generated result yet, so use it for timing and framing only.
Step 5: Check it, then post it
Watch it once with the sound on, then pause on each of the eight turns. Check that each person is the right person, then watch the two seams, the hands near the lens and the mouths on the chorus. The result checklist covers each check and when to re-run.
Then post it with a caption that names the group, add "wait for the last one" on screen and send the file to all eight. There is more in captions, cover and who to cast last.
Say that the video is AI-made, and check how your platform handles commercial music before posting for a brand.
Related reading
- The Baby Lip Sync trend explained: why the format works and how the template rebuilds it.
- The Justin Bieber Baby AI trend: the song and what this template does not do.
- Who sings when: every turn in the 37 seconds.
Frequently asked questions
How many photos do I need for the Baby Lip Sync video?
Eight, one for each of the seven singers and one for the closer. All eight are required.
Can I use one person more than once?
You can use a photo twice, but that person will then appear twice in the same room, sometimes in the same frame.
Do we need to record ourselves singing?
No. The video keeps the clip's original audio, and the mouths follow the reference performance.
What shape and length is the video?
About 37 seconds, portrait 9:16, as one long moving take.
Is the preview video made by Oakgen?
No. It is the reference clip the trend comes from. It is not an Oakgen result.
