Hotel Lobby AI prompt (copy and paste) for the orange wall rap video
The copy-ready Hotel Lobby AI prompt: two people from two photos rap at one hanging mic in front of an orange wall, beat by beat. Pet version and fixes.
7 最小読み取り · 更新済み
写真2枚でHotel Lobbyの主役に
作ってみるThe Hotel Lobby AI video prompt
Here’s the copy-ready prompt for the Hotel Lobby trend: two people at an orange wall rapping at one hanging silver mic, with the open palms, a full spin, a hush to each other and a hand shading the eyes, in one widescreen shot. Attach two photos as reference images, the one on the left first, and paste it in.
Two people
A widescreen 16:9 rap performance video, exactly 15 seconds, photoreal live action in full natural colour, one continuous locked-off shot, no cuts. Two people stand side by side, framed from the waist up. No text, no subtitles, no logos, no watermarks. Setting: a plain, evenly lit orange wall fills the whole background edge to edge, and exactly one silver vintage microphone hangs on its cable from above, between them, at mouth height. Nobody holds it. Nothing else is there: no other people, no props, no furniture. WHO: the person on the LEFT of the frame is the subject in @Image1, the person on the RIGHT of the frame is the subject in @Image2. They never swap sides. Each wears exactly the clothes from their own photo. ACTION, on the beat; both rap along the whole time, mouths moving: [0-4 s] LEFT looks up at the hanging microphone and raps toward it, hands clasped at the chest. RIGHT raps too, reaching one hand toward the microphone, then tapping the wrist of the other arm. [4-7 s] RIGHT spreads both hands wide, palms open, then clasps them at the chest. LEFT nods to the beat. [7.5-11.5 s] RIGHT raises one arm high, turns all the way around so the back faces the camera, then turns back to the front. LEFT keeps rapping toward the microphone. [12.3-12.8 s] Both turn to each other and hush, one finger to the lips. [14-15 s] RIGHT holds a flat hand above the eyes, shading them and looking into the distance. IDENTITY: LEFT is exactly the person from @Image1 and RIGHT exactly the person from @Image2 in every frame: the same face, age, skin tone, hairstyle and hair colour, the EXACT body weight and build, and the exact clothes each wears in their photo. They never swap faces, sides or clothes. Avoid: a second microphone, a hand-held microphone, a mic stand, a patterned or red background, a third person, cuts, zoom, slow motion, face changes, swapped faces, captions, text, logos, watermarks, extra arms, extra fingers, distorted hands, morphing faces.
A prompt gives the tool a plan, not a promise; two or three runs is normal. Want it without a prompt, with the song in? The Hotel Lobby page makes it from two photos.
How to use it
- Open an AI video tool’s reference-image mode that takes two images: Kling, Higgsfield, Runway, Veo or CapCut’s AI video have one.
- Attach the photo of the one on the left first, then the one on the right.
- Paste the prompt. If your tool doesn’t use @Image1 and @Image2, swap them for “the person in my first reference photo” and “the person in my second reference photo”.
- Set 16:9 and 15 seconds, or the nearest length. Turn off the tool’s own music if it adds some.
- Generate, watch it against the frames below, and use the fix-it table for anything off.
What each part of the prompt does
The first line
Widescreen, 15 seconds, one locked-off shot with no cuts, both framed from the waist up. That’s the look of the original performance clip: the camera never moves, and all the fun is in what the two of you do.
Setting
A plain orange wall edge to edge and exactly one silver mic hanging between you that nobody holds. “Nobody holds it” matters: without it, tools hand the mic to one of you or add a stand.
WHO
LEFT is the subject in @Image1, RIGHT the subject in @Image2, and they never swap sides. Naming them by side lets the ACTION lines say who does what.
ACTION
Five timed lines, one per beat: rapping at the mic, the open palms, the spin, the hush and the lookout. “Both rap along the whole time, mouths moving” keeps it from turning into a silent dance.
IDENTITY
With two people, the usual problem is faces and clothes swapping between them, worst of all in the spin. This line keeps each person themselves in every frame, with the same face, hair, build and clothes.
Avoid
The things video tools tend to get wrong in this clip, named so they’re less likely: a second or hand-held mic, a mic stand, a different background, a third person, cuts and zooms, swapped faces, text and the usual hand problems.
There’s no sound line. We make the video silent and lay the HOTEL LOBBY song over the whole 15 seconds afterwards, so do the same: add the song when you post.
Beat by beat: what each line does
- The mic: “looks up at the hanging microphone” makes them rap at it, not at the camera.
- Open palms: “then clasps them at the chest” brings the hands back, so the next beat starts clean.
- The spin: “so the back faces the camera, then turns back to the front” forces a full turn, not a quarter step.
- The hush: “turn to each other” is what makes it a duo moment; facing the camera it loses the joke.
- The lookout: “a flat hand above the eyes” is the last frame people remember, so it gets its own line.
Variants
Two pets
Change the WHO line to “the pet on the LEFT is the animal in @Image1, the pet on the RIGHT is the animal in @Image2, both standing upright on their hind legs, paws free”, and change “both rap along” to “both open and close their mouths to the beat”. Keep the rest. Attach one photo of each pet, whole body, at its eye level.
You and your pet
Put yourself as @Image1 on the left and your pet as @Image2 on the right, standing on its hind legs. Give the spin to yourself if your tool keeps getting the pet’s turn wrong: swap LEFT and RIGHT in that one line.
A different wall
Want a pink, teal or yellow wall? Change “orange” in the Setting line and keep “plain, evenly lit” and “edge to edge”. The hanging mic is what makes it read as Hotel Lobby, so keep that.
Shorter clip
If your tool only makes 5 to 10 seconds, keep the mic, the palms and the hush, and drop the spin and the lookout. The hush makes a good last frame.
Fix it: symptom and change
| What went wrong | Change in the prompt |
|---|---|
| The two of you swapped sides | Keep “They never swap sides” and attach the photos in the right order |
| A face changes during the spin | Keep “in every frame” in the IDENTITY line and use sharper photos facing the camera |
| Someone holds the mic | Keep “Nobody holds it” in the Setting line and “a hand-held microphone” in Avoid |
| Two mics, or a mic stand | Keep “exactly one silver vintage microphone hangs on its cable” and both in Avoid |
| The camera cuts or zooms | Keep “one continuous locked-off shot, no cuts” and “cuts, zoom” in Avoid |
| Only a half turn | Keep “turns all the way around so the back faces the camera, then turns back” |
| Mouths don’t move | Keep “both rap along the whole time, mouths moving” above the ACTION lines |
| Someone looks slimmer or younger | Keep “the EXACT body weight and build” in the IDENTITY line |
The photos you attach
- One face per photo, facing the camera, sharp and well lit, nothing covering it.
- Head to knees is good: you’re framed from the waist up, in the clothes from your photo.
- No sunglasses; the faces are close and in frame the whole time.
- No group photos, and only your own photo or an adult’s who has said yes. No celebrities and no one under 18.
Examples are in the photo guide.
What a prompt can’t do
- Add the song. You pick HOTEL LOBBY as a sound when you post.
- Promise both faces stay right through the spin. Check each run.
- Promise the same take twice. Every run is a new performance.
The no-prompt route: add two photos on the Hotel Lobby page and get the finished video with the song in. If it fails, the credits come back automatically, and prices are on Pricing. Comparing options? See CapCut template vs AI vs two photos.
NowStarring is an independent fan tool, not linked to Quavo, Takeoff, their label or COLORS, and the song belongs to its rights holders.
よくある質問
What is the prompt for the Hotel Lobby trend?
Describe a widescreen 15-second single shot: two people at waist height in front of a plain orange wall, one silver mic hanging between them, rapping, then open palms, a full spin, a hush to each other and a hand shading the eyes. The full copy-ready prompt is at the top of this guide.
Does it work in ChatGPT, Gemini or Higgsfield?
It works in any video tool that takes two reference images and a text prompt, including Veo in Gemini, Higgsfield, Kling and Runway. A chatbot that only makes images won’t make the video.
What do @Image1 and @Image2 mean?
They’re the two photos you attach, in order. If your tool doesn’t use @Image1, write “the person in my first reference photo” and “the person in my second reference photo”.
Which photo goes first?
The person on the left goes first, as @Image1. The one on the right goes second, as @Image2, and does the palms, the spin and the lookout.
Will the prompt add the song?
No. There’s no sound line in it: we make the video silent and lay the HOTEL LOBBY song over it afterwards. Pick the sound on TikTok or Reels when you post.
Is there a prompt for cats or dogs?
Yes. The two-pet variant below swaps the people for two pets standing on their hind legs, one from each photo.
