If you have opened TikTok, Instagram Reels, or YouTube Shorts recently, you have undoubtedly run into it: two characters—whether historical figures, popular meme characters, or even pets standing upright—standing side by side in a bright orange studio booth, rapping and nodding along to Quavo and Takeoff's iconic track "Hotel Lobby".
Inspired by the distinct visual staging of the COLORS music show, this AI video trend has become a massive social media phenomenon. The best part? You do not need professional 3D rigging or after-effects software to create your own version. Here is the step-by-step breakdown of how the meme works and the exact prompt architecture you need to replicate it.
Step 1: The Foundation Requires Two Subjects
Most AI dance and lipsync templates are engineered for a single face. The Hotel Lobby trend works because it captures dynamic contrast between two individuals trading lines. To generate a high-performing video, your source image must feature:
- Two full-body or half-body subjects standing next to each other.
- A clean, flat background (ideally studio orange or minimalist monochrome).
- Direct eye contact with the camera.
Step 2: The Core Prompt Architecture
When feeding your reference image into text-to-video or image-to-video engines, camera stability and sequential movement timing are critical. Use this structured prompt framework as your foundation:
Master Prompt:
"Cinematic vertical 9:16 music-video shot, locked static camera, no cuts. Two subjects standing inside a monochromatic orange studio booth with a single microphone suspended from the ceiling between them. The left subject leans toward the mic, rapping with expressive facial cadence and subtle hand gestures on the beat. The right subject bounces rhythmically in the background, making direct eye contact with the camera. Realistic fabric folds, soft studio lighting, sharp focus, 4k resolution."
Step 3: Reverse-Engineering Your Favorite Variations
Because creators are constantly modifying the trend—adding custom camera glides, neon studio lighting, or surreal art styles—guessing the exact text prompts used in viral clips can lead to wasted credits and inconsistent renders.
The fastest way to replicate a style is to extract the cinematography data directly from the clip. Using VromptAI, simply upload the video file of the variation you want to emulate. VromptAI's vision engine breaks down the clip's camera motion, lighting setup, and stylistic tokens into an editable, production-ready text prompt that you can plug directly into your preferred video generation model.