Hotel Lobby AI Tutorial for Desktop: Two Photos to a Viral Video
A desktop walkthrough of the Hotel Lobby AI generator: which photos to pick, how the two upload slots work, and how to get a clean orange-booth video on the first try.
This tutorial covers the desktop flow of the Hotel Lobby AI video generator, end to end. The whole process takes a couple of minutes once you have the right photos.
Before you start: pick your two photos
The generator casts two performers — the left slot and the right slot of the orange booth. Everything about the result depends on the two photos you choose:
- One subject per photo. A group shot gives the model one crowded read and the faces blur. Crop first.
- Front-facing, evenly lit. Unobstructed faces keep both identities recognizable.
- Full-body if possible. The booth frames both performers head to toe, so visible feet help the placement.
One person can also play both roles — just upload two different photos of the same person. Different outfits or expressions make the duet funnier.
Step 1: Open the generator
Go to the Hotel Lobby AI video generator. On desktop you will see the two upload slots on the left and the booth preview on the right.
Step 2: Fill the two slots
Click the first slot and upload the photo of the performer who should stand on the left — they open the performance. Click the second slot for the performer on the right. You can also drag and drop image files onto the slots. JPG, PNG, and WebP up to 8 MB work.
Your photos stay in your browser during this step — they are used for the preview and are not published anywhere.
Step 3: Generate and preview
Click Generate. The template already contains the orange studio, the hanging microphone, and the two-person framing — there is no prompt to write and no editing timeline. One performer raps while the other nods and reacts, then they switch, just like the original format.
Step 4: Download and post
The result is a vertical 9:16 clip, ready for TikTok, Instagram Reels, or YouTube Shorts. When you post, add the “Hotel Lobby” track through the platform’s licensed sound library so the audio is cleared.
Troubleshooting
- Faces look blended? You probably used one group photo. Re-crop into two separate photos.
- A performer is cut off? Use a full-body photo with some space around the subject.
- Wrong side? Swap the two upload slots — the first photo always takes the left.