AI Carpool Karaoke
Turn one photo into a silent car sing-along performance: the people appear to sing, sway and laugh in the front seats of a car.
Use one photo of one or two people with clear faces and visible upper bodies. It is an identity reference for a newly generated 16:9 car scene. The output is a silent visual only: no music, vocals, generated audio or song-specific lip-sync. Add your soundtrack afterward.
568 credits per 5-second video
External scene references, not VideoAny-generated results. Your 5-second clip may differ.
Explore the car sing-along mood
These external references show front-seat framing and playful performance gestures that inspire the silent visual effect.
Front-seat duet reference
Two adults appear to sing from the front seats, viewed from the dashboard.
SuperMaker ↗
Daylight car performance reference
Two adults share a relaxed front-seat performance in warm daylight.
SuperMaker ↗External scene references, not VideoAny-generated results. Your 5-second clip may differ.
Upload your photo
Choose one clear photo of one or two people with faces visible.
Generate your video
Sign in and check that you have 568 credits. Select Generate to start your 5-second clip.
Preview and download
When the video is ready, watch the result and download it. Faces, motion and backgrounds can vary.
AI Carpool Karaoke FAQ
It generates a silent visual performance with the people from your photo in the front seats of a car, moving their mouths as if singing, swaying and laughing. It does not generate music or singing audio.
Yes. Use one photo containing one or two people. With one person, the prompt places only that person in the driver’s seat. With two, it assigns them the driver and passenger seats without adding another person. The photo is an identity reference, not a fixed starting frame; AI details can vary.
No. Audio generation is disabled for this effect. It creates silent mouth movements and gestures without a song or voice. Add your own soundtrack afterward; the motion is not synchronized to a particular track.
Each generation requires 568 credits. If you have fewer than 568 credits, add credits to continue.
No. This is an independent effect inspired by the car sing-along format and is not affiliated with any TV show or network.
Image quality, content checks and temporary service issues can affect generation. AI may change facial details or the background. If progress cannot be checked, review your generation history before starting another task.
Each generation uses Seedance 2.5 to create a 5-second, 720p landscape 16:9 video at 568 credits. Five seconds is the video length, not the generation time.
Ready to create your next release?
Upgrade for faster queues, higher resolutions, and expanded usage options.