Two faces, one shared mic
Hotel Lobby Video turns two portraits, or one photo of you both, into an AI duet on the orange set. A 12-second take costs 6 credits and is usually ready in about 5 to 15 minutes.
Your two faces
A single face per picture. Left leads, right backs up.
Set
Take and sound
Price: 6 credits. Done in roughly 5 to 15 min; if it fails, the credits come back. See prices
Around the trend
A set list of Hotel Lobby remakes, explainers and how-tos that other people posted. Press play and each one runs right here in the platform's own player, credited to its maker. None of them were made with Hotel Lobby Video.
On stageTwo realistic AI cats at the micby AI Motion LabOpen on YouTube
Videos play from YouTube or X and stay the work of their creators. We link every original post; ask us at support@hotel-lobby-video.app if you want yours removed.
One session, three moves
Treat it like a short recording session: cast the pair, pick what plays, then let the studio cut the take.
Drop in a portrait for each face, or tap One photo if all you have is a picture of the two of you together. Whoever is on the left takes the verse; the right side handles the ad-libs. Faces in good light and outfits in frame give the model more to copy.
Wan 3.0 keeps the original Hotel Lobby track or plays a section of your own audio. Switch to a Seedance model and type a topic when you want a brand new verse instead. Nothing to script: the routine, the set and the camera are already built in. The create button shows length, quality and the exact credit count before anything is charged.
Once the take is rendering you can close the tab; it waits for you in My videos. Play it through, check both faces, then save the MP4, which carries no watermark. If you get stuck, the full walkthrough goes through each choice slowly.
Sample clips
These clips were made with Hotel Lobby Video. Select a sound button to hear the full take.
What a take costs
On Wan 3.0 at 480p, a 12-second take is 6 credits. The smallest pack is 10 credits for $9.99, pack credits never expire, and plans are optional. Longer takes, sharper quality and other models raise the count.
$9.99 paid once
$5.99 for each 12 s take at 480p
Payments open shortly
$29 paid once
$3.48 for each 12 s take at 480p
Payments open shortly
$59 paid once
$1.77 for each 12 s take at 480p
Payments open shortly
$99 paid once
$1.70 for each 12 s take at 480p
Payments open shortly
$199 paid once
$1.49 for each 12 s take at 480p
Payments open shortly
Payments handled by Stripe · Pack credits stay valid forever · Plans renew until you cancel · A failed take costs nothing
A credit is not a dollar. Adding sound costs nothing extra, though another model may change the price. The create button always shows the exact total for your choices. If a take fails, its credits come back to you. See every price
The mixing desk
Every take starts from the Hotel Lobby routine; what you control is the soundtrack and the file you get back. Wan 3.0 plays the original track or audio you upload, while the Seedance models write and perform a new verse. The studio only lists the sound options that fit the model you picked.
On Wan 3.0 you choose between the Hotel Lobby original, an audio file of your own (MP3, WAV or M4A all work) or the sound of a reference clip. Trim your audio to 2 to 15 seconds right in the browser, and only use audio you have the rights to. The own-song page covers voice notes and recorded messages too.
Pick a Seedance model and write what the rap should be about, up to 120 characters. The model writes and performs new vocals over a new beat, so neither the Hotel Lobby recording nor any upload survives. Both rap starts switched on and has the pair trade lines; switch it off for one front voice and a sidekick. More on the AI rap generator.
Takes on the Hotel Lobby set run 5, 8, 10, 12 or 15 seconds. Wan 3.0 renders 480p, 720p or 1080p, with 480p preselected. Go 9:16 for phone feeds or 16:9 for a laptop or TV. Upload a short dance clip of your own and the pair follows its moves, though not frame for frame, and the dancer’s face is never copied.
Choosing a sound costs nothing by itself, but a sound that needs a different model changes the price. A reference clip of 2 to 15 seconds is rounded to a whole-second take between 4 and 15 seconds. The final length and credit count sit on the create button.
Every take is paid for with credits. Packs from $9.99, no subscription. See example takes or check prices.
Who takes the mic
Two roommates, a pair of coworkers, the friend who moved abroad: separate portraits are fine, you never need a photo taken together. Settle who gets the verse first, because the left face leads.
Pairs across generations are often the funniest. Pick recent, sharp photos where you can; faded prints and blurry scans give the model less to hold on to, so details may drift. Ask both people first, and get a guardian’s okay for anyone under 18.
Aim the AI rap at the birthday person, or upload a voice note of the family singing and let the lead lip-sync it. Plan for a second take, since a generated performance does not always land the first time. More ideas on the birthday rap page.
Bring a track you are allowed to use, or a short clip of a dance, and the pair performs it on the orange set. Change one thing per take so you can tell what made the difference. Each new take costs credits; changing a setting never reworks a finished video.
Prep your portraits
The model can only copy what it can see, so sharp, bright faces matter most. Even then, faces, outfits, gestures and lip-sync shift from one take to the next. The example clips are picks, not a promise that every pair of photos turns out the same way.
Good picks
Avoid these
All pages
Each page opens Hotel Lobby Video with a matching preset. Pick an idea, then add your portraits.
A plain definition of Hotel Lobby AI (AI remakes of a 2022 COLORS rap set, recast with two new people), why it is not a camera filter, and how Hotel Lobby Video stages two people from portraits
Start thisUpload two portraits and get a rap duet of those two people on the orange COLORS-style set
Start this
The duo’s COLORS set restaged with a portrait of you and one of a friend, one voice on the verse and one on the ad-libs
Start thisThe front voice lip-syncs 2 to 15 seconds of audio you upload, such as a demo, a voice memo or a birthday shout-out
Start thisA one-line topic becomes a fresh, clean verse and beat, rapped on camera by the two people in your photos
Start thisYour portrait and a friend’s become a rap duet with a fresh verse, front voice and sidekick or both trading lines
Start thisOne sentence and a pair of portraits become a fresh rap with a beat, performed on camera by those two people
Start thisA clean, fresh diss rapped by two people from your portraits, a one-way roast or a back-and-forth
Start thisA pair of portraits and one argument become a video of those two people trading fresh, clean bars at one mic
Start thisA personal birthday rap written about one person and performed by them and a friend, ready to send on the day
Start thisThe birthday person and a friend perform a clean roast rap about them, or one portrait gets a comic photo effect
Start thisYour own track lip-synced, or an AI-written rap, performed on one of 4 sets
Start thisThe two people in your portraits perform a rap together on the orange Hotel Lobby set or three others, with an AI rap or your own track
Start thisYour dog, cat or other pet on the orange Hotel Lobby set from two photos, at the single mic with you or with another pet
Start thisCan ChatGPT make videos? In October 2026 it produces text and still images, not video, and Sora has shut down; use ChatGPT for the idea and Hotel Lobby Video for a video of you and a friend from photos
Start thisThe owners’ photos and one line about the shop become a short, fresh rap ad that they perform themselves
Start this28 ready-made AI scenes that turn a still picture into a 5 second clip, 4 credits at 720p
Start thisA single selfie becomes one unbroken camera climb from your face, over the rooftops, through the clouds and out to the globe
Start thisTwo pictures taken in different places become one short clip of the pair facing each other, smiling and holding on in a long embrace
Start thisA single picture becomes a short clip where copies of you join one at a time until the whole crew lands the same pose
Start thisA single picture, or a picture per person, becomes a slow sung moment at the end of a wooden jetty, sun going down, warm drizzle in the light
Start thisTwo pictures become a short, foggy street scene at dusk where the first person slowly turns into a zombie and the second one stays and hugs them
Start thisA single picture becomes a three-cut sunset edit at a deserted fuel stop, with a dance between the pumps, a grin into a selfie and an exit from a black luxury car
Start thisA picture, or a picture per person, becomes a short house-party clip where a swirl of orange and violet sparkles turns everyday clothes into a homemade-style costume
Start thisOne picture, or a picture per person, becomes a short head-to-toe routine on a glossy neon studio floor, in the clothes you wore, set to an original dance track
Start thisA picture, or a picture per person, becomes a short prank clip where all the hair disappears with a cartoon pop and the newly bald heads crack up
Start thisA picture, or a picture per person, becomes a short camcorder home movie from around 1985, with big hair, retro clothes, a sunny suburb and an original synth-pop tune
Start thisToday’s you and an earlier you stand inside a white instant-film border, turn toward each other and hold on, with a music-box melody
Start thisYour portrait and a portrait of a loved one who has died become a short, imagined clip of the two of you embracing in a golden field, scored with quiet piano and strings
Start thisA single portrait turns into a short clip in which the person blinks, breathes, smiles and turns their head slightly, inside the photo’s own background, with quiet piano
Start this
Short clips from family photos, either their own portrait gently coming to life or an imagined hug between you, ready to place in a slideshow for a service
Start thisUse one portrait to make an AI video of yourself in a browser, compare solo scenes with two-person options and see the price before creating
Start this
A browser-based way to make a TikTok AI trend clip from a photo, with effect choices, credit costs and steps for adding a sound after export
Start this
A free rap duo name generator that turns two first names, one name or an empty form into 8 clean options to consider
Start thisAudio included
Hotel Lobby AI With the Original Song
Your choreography, their faces
Hotel Lobby AI With Your Own Dance Video
The smallest voice on the mic
Hotel Lobby AI With a Baby Photo
Your face stays yours
Hotel Lobby AI That Keeps Your Own Face
One shot, two performers
Hotel Lobby AI With One Photo of Two People
HOTEL LOBBY · HOW TO
3
Hotel Lobby: three ways
HOTEL LOBBY AI PROMPT
Direct the performance
THE SONG'S STORY
2022 → 2026
Hotel Lobby: from song to trend
FAMILY & MUSIC
Quavo & Takeoff
COMPARE YOUR OPTIONS
Pick your Hotel Lobby tool
FUNNY RAP LYRICS
18
Clean bars. Everyday chaos.
FRIENDLY RAP BATTLE ROASTS
25
Roast the habit. Keep the friend.
AI RAP VIDEO · HOW TO
3
From photos to a rap video
YOUR OWN SONG · TWO SINGERS
2
One song, two singers
DOG RAP VIDEO · HOW TO
3
Your dog, on the mic
THE BEST RAP DUOS
16
Two voices. A shared legacy.
RAP SONGS & FRIENDSHIP
9
The soundtrack to friendship
AI RAP GENERATORS
What should AI make for you?
AI MUSIC VIDEO TOOLS
Three routes to a music video
HOTEL LOBBY · DUO IDEAS
15
The duo makes the story
HOTEL LOBBY AI · FREE?
Free… up to which step?
HOTEL LOBBY · CAPCUT
Template or AI performance?
HOTEL LOBBY · HIGGSFIELD
Motion transfer, explained
THE MIGOS AI VIDEO TREND
One scene. Unexpected duos.
HOTEL LOBBY AI · APPS
Which app makes what
HOTEL LOBBY AI · ON YOUR PHONE
Make it on your phone, no app
BIRTHDAY RAP VIDEO · HOW TO
A birthday rap from two photos
VOICE NOTE · HOTEL LOBBY AI
Your voice note, their faces
HOTEL LOBBY AI · PHOTO GUIDE
Pick photos that keep the face
HOTEL LOBBY AI · YOUR OWN MOVES
Your clip, their faces
HOTEL LOBBY AI · OUR TEST LOG
5
What went wrong, what we fixed
AI RAP BATTLE · HOW TO
Two photos, one rap battle
AI RAP VIDEO · CAPTIONS
Add captions, fix the names
HOTEL LOBBY AI · RESOLUTION
480p, 720p or 1080p?
AI RAP · ANY LANGUAGE
An AI rap in your language
Our own takes
Two sample renders, with model, length and source shown under each. They demonstrate the workflow, not a promise about how often a take succeeds.
Wan 3.0 · 480p · 10 s · 16:9
Wan 3.0 · 720p · 12 s · 16:9
Where the set comes from
Hotel Lobby trend videos are AI remakes of Quavo and Takeoff performing their song “Hotel Lobby” on A COLORS SHOW in 2022: a plain orange wall, a single mic hanging between two people, one carrying the verse while the other hypes. From September 2026 people started casting their friends, partners, grandparents and pets in those two spots. The same clips also circulate as Migos AI or the AI rap duo trend.
Hotel Lobby Video is independent, with no tie to the artists, their label or COLORS. Only use faces and audio you have permission to use.
How the trend started and spread →No. There is no free take and no free preview; even your first video is paid with credits. The smallest pack is 10 credits for $9.99 and needs no subscription. You can watch the example clips on this page before you buy anything.
Yes. Open the site in your phone’s browser, tap One photo and add the picture; whoever stands on the left takes the verse. Buy a one-time pack, which never expires, and no plan is needed. Both faces should be clear and roughly the same size. More on one photo of two people.
Yes, with the default setting on Wan 3.0: the MP4 already has the Hotel Lobby recording in it. Having it in the file is not a license, so for a public post the platform’s own Hotel Lobby sound is the safe choice. The Seedance models replace it with a new AI rap. See the original-song page.
Yes, on Wan 3.0. Upload the audio and keep a 2 to 15 second section; the lead lip-syncs it. The site does not record in the browser, so record the voice note in your phone’s recorder first and upload the file. How closely the lips follow the audio varies. Details on using your own song.
Yes. A pet photo works in either spot, next to you or next to another pet. We have tested a dog next to a person; cats and two pets together have not been tested yet. Use a sharp photo where the face is fully visible. See the pet rap page.
No. Every duet has exactly two performers, from two portraits or one photo of both. With three people, make two takes with different pairs, or pick a photo effect, which takes one or two people.
Our own Wan 3.0 takes needed about 5 to 15 minutes. Treat that as a guide, not a promise: queues, length and quality change the wait. After the take starts you can close the tab and find it later in My videos.
There is no watermark. On iPhone, Safari saves the MP4 to the Downloads folder in the Files app; open it, tap Share, then Save Video to move it to Photos. On Android, Chrome puts it in Downloads. Captions are not burned in, so add them in TikTok, Instagram or CapCut.
480p is the default and the cheapest. Instagram asks Reels for at least 720 pixels and 30 fps, and a 720p vertical take on Wan 3.0 is 720 × 1280 at 30 fps. 1080p costs twice 720p. A side-by-side comparison is in the resolution guide.
Not always. AI can shift a face, an outfit, a gesture or the lip-sync, and moves from a reference clip can carry over. Sharp photos improve the odds; watch the whole take before you share it. A take that rendered successfully can still look different from what you had in mind.
A failed take puts its credits back in your balance automatically. This is a credit return, not a cash refund; cash refunds follow the refund policy. A finished take you simply do not like is not refunded, and another attempt uses new credits.
No. The trend has no official tool, and Hotel Lobby Video is not affiliated with the artists, their label or COLORS. It is also not a phone app: it runs in any modern browser on a phone or a computer, and you sign in with Google.
Uploaded photos, songs and clips held by our service are scheduled for deletion 7 days after the take finishes. Finished videos stay in your account until you delete them or close the account. The AI providers that process a take keep data under their own terms. Read the privacy policy.
No. Ask everyone whose face you upload, get a guardian’s consent for minors, and never use a real person to make it look as if they said or did something they did not. A yes to the photo does not cover the song: music rights are a separate question, and you are responsible for both.
The studio is open
Cast two faces, choose what plays, and check the credit count on the button before you start the take.
