The video supplies the performance
Keep the source’s actions, camera, stage and timing. The default already includes this direction, so you do not need to describe every gesture.
SAME PERFORMANCE. YOUR CAST.
The familiar duet. Your faces, hair and outfits.
Upload two photos to replace both performers in the orange studio. The original clip guides the actions and camera; its soundtrack is kept in the export.
Photo 1 → left performer · Photo 2 → right performer. Original duet keeps the source audio. Generation uses credits; review the cost before submitting.
WHY THIS CLIP IS RECOGNIZABLE
The Hotel Lobby AI trend reimagines Quavo and Takeoff’s HOTEL LOBBY performance: two performers, a hanging microphone and a bright orange studio. Often searched as the Migos AI video trend, its recognizable actions and music are part of the joke when friends, grandparents or characters take the performers’ places.
Original duet uses the actual reference video alongside your photos. The video defines how the performance moves; each photo defines who the performer looks like. This replaces faces, hair and clothes together.
TWO PHOTOS TO YOUR VERSION
Use one clear person per image. The first photo replaces the left performer; the second replaces the right. Include hair and the clothes you want to appear in the video.
The reference clip is included automatically in Original duet mode. Its actions, camera, orange studio and timing guide generation. Add an optional description to refine the replacement appearance.
Start with a 5-second opening segment, or choose 10 seconds with an eligible paid plan. The source audio is copied into the exported clip. Check both identities, hands and lip-sync before sharing.
NO LONG PROMPT NEEDED
Keep the source’s actions, camera, stage and timing. The default already includes this direction, so you do not need to describe every gesture.
Use a clear photo of each person, including their hairstyle and outfit. Keep the upload order consistent: left first, right second.
For example: “Keep the first person’s curly hair and red jacket, and the second person’s glasses and white shirt. Keep both faces visible.” Changing the scene or camera changes the reference look.
EXPLORE OTHER CASTS
These earlier examples show new orange-studio scenes, including grandparents and pets. Their soundtrack and tutorial captions were added in editing. They illustrate cast ideas, rather than demonstrate the current reference-based duet pipeline.
A two-person orange-studio performance with tutorial captions added after generation.
10 seconds · Captioned tutorialAn original older couple brings a relaxed rhythm to the mic.
8 seconds · Original AI exampleA tabby and a gray cat trade playful paw gestures.
8 seconds · Original AI exampleA cream cat and a corgi share the orange studio.
8 seconds · Original AI exampleBEFORE YOU TAKE THE MIC
Also called the Migos AI video or AI rap duet trend, it puts new performers into the recognizable orange studio and hanging-microphone performance associated with Quavo and Takeoff’s HOTEL LOBBY. The appeal is the familiar performance with an unexpected cast.
Original duet replaces the full appearance of both performers: face, hair, body appearance and clothing. The two photos supply the new cast; the reference video supplies movement, camera, scene and timing. This uses generative video editing, so individual frames and lip-sync can vary.
No. Original duet includes the reference clip automatically. Upload your two photos and generate here. A description is optional; use it for visual adjustments rather than rewriting the choreography.
Original duet copies the reference clip’s existing audio into the export instead of asking AI to sing or recreate it. A 5-second result uses the opening segment. Original audio is fixed in this mode; Solo and Pet duet generate new performances and do not use this soundtrack.
Original duet needs exactly two photos, with one clear person per photo. Upload the left performer first and the right performer second. Avoid group photos, covered faces, and heavy blur. Solo needs one photo; Pet duet needs two pet photos.
Yes. Select Pet duet or Solo in the performance settings. These options generate a new scene inspired by the orange studio, rather than editing the original two-person performance. Their movement and audio differ from Original duet.
The default is 16:9 to keep the reference composition. You can choose 9:16 for Shorts or Reels, but reframing changes the shot and may crop the performers. Review both subjects before posting.
Choose 5 or 10 seconds. The supplied reference is 10 seconds long. The default is 5 seconds; 10 seconds requires an eligible paid video plan. Check the credit cost before submitting.
Generation uses Vismuse credits. Review the cost shown before submitting and see the pricing page for current subscriptions and one-time purchase options. Revisions can use additional credits.
The reference constrains motion, camera and timing, and the source audio is reused directly. The picture is still generated: exact choreography, identity consistency and perfect lip-sync are not guaranteed. Review the result before sharing.
See Vismuse pricing for current credit options.
Bring two clear photos. Keep the reference performance and give it a new cast.
Upload your photos