Side AMigos AI with my own song
Hotel Lobby AI, Lip-Synced to Your Own Song
Yes, you can trade the Hotel Lobby track for your own audio. Tap Your song, upload an MP3 or WAV, choose the 2 to 15 seconds you want, and the person in your left photo performs it into the hanging mic while the right one hypes.
- A1The Your song button
- A2MP3 or WAV, any length
- B1Pick 2 to 15 seconds
- B2Wan 3.0 only
Who’s on the record
Stage
Video & sound
This video costs 6 credits and takes about 5 to 15 minutes. If it fails, your credits come back. See prices
Liner notes · The short answer
Yes. On MigosAI (migosai.si) you can make the Hotel Lobby AI video, also known as the Migos AI video, with your own song: add two photos, choose Your song as the sound, upload an MP3 or WAV and pick a 2 to 15 second part in the browser. The left person lip-syncs it into the hanging mic and the right person hypes. Uploads work on Wan 3.0 only; the Seedance models generate their own AI rap instead. Your audio never changes the price: 6 credits for 12 seconds at 480p, packs from 10 credits for $9.99, no subscription needed. The render usually takes about 5 to 15 minutes.
Checked against the generator on
How it works
What happens when you pick Your song
A Hotel Lobby AI video on MigosAI has three inputs: two photos for the people, the booth routine for the moves, and one sound for the lips. Your song replaces only that last input. The orange booth, the hanging mic and the routine stay exactly as they are.
Your file is trimmed on your own phone or computer before anything is sent. Drag the orange frame to the part you want and play it back to check. The frame is as long as the video, so a 12-second video takes 12 seconds of your song. If the file is shorter than the video, the last seconds simply have no vocals.
Only Wan 3.0 follows an uploaded track. If the generator is on a Seedance model, it offers to switch you back, because those models write and sing a new AI rap and ignore uploads. Use audio you made or have the right to use: your own beat, a voice memo, a toast you recorded.
Can two people sing different lines of my song?
Not in one video. An uploaded song is one voice track, and the person on the left is the one who lip-syncs it; the person on the right reacts. If you each want a turn, render two videos, swapping the photos and picking a different stretch of the song each time, then join them in any editor. Every step is in two people, one song. If you want two voices trading lines and new words are fine, use AI rap with Both rap on a Seedance model instead.
How exact is the lip-sync?
Good enough for a short social clip, not studio-perfect. Mouths track best when the vocal is loud and clear over the beat, when one voice sings at a time and when your part starts on a word rather than on silence. Heavy reverb, stacked harmonies, crowd noise and very fast double-time bars give the model less to follow, and the mouth can drift.
Step by step
How to use your own song in the Hotel Lobby booth
- A1
Photos first
Left photo: the one who sings your track. Right photo: the hype partner. A single picture of both of you works with One photo.
- A2
Upload under Your song
Open Soundtrack & moves, tap Your song and choose the file: MP3 or WAV, and an M4A voice memo works too. Drag the orange frame to your part and press play to hear it.
- B1
Set the length, then Create
In Video settings pick 5, 8, 10, 12 or 15 seconds; the frame follows that length. Press Create: 6 credits for 12 seconds at 480p, the same as with the original track.
Ideas
What to put in the booth besides “Hotel Lobby”
- A1
A song you wrote
Your hook, performed by you and your producer, as a teaser before release day.
- A2
A recorded toast
A short wedding or birthday line over a beat, delivered by the couple or the birthday girl.
- B1
The group-chat voice memo
That one voice note your friends still quote, finally given a stage and a hype man.
- B2
A verse in another language
Lip-sync follows the audio, not a script, so a verse in Spanish, Tagalog or Polish can be used too.
Own song: what you can and cannot do
| You can | You cannot, and what to do instead |
|---|---|
| Upload an MP3 or WAV of any length | Use more than 15 seconds in one video: make two and join them |
| Choose exactly which part plays | Have both people sing different lines: swap the photos between two videos |
| Keep the booth routine on Wan 3.0 | Use your upload on Seedance: those models write their own AI rap |
| Pair your song with your own dance clip | Fix muddy audio in the tool: upload a clean recording instead |
| Pay the same as any video of that length | Get a free test render: every video is paid in credits |
A quick check before you upload
- Is it yours, or do you have permission? Commercial songs belong to their labels.
- Does your part start on a word? A quiet intro leaves the lead nothing to mouth.
- Is one voice singing at a time? Duets and choirs make it unclear whose lips should move.
- Is the part at least 2 seconds long? Anything past 15 seconds is left out.
Rather have the AI write the words? The AI rap generator makes a new verse from a topic, and the voice note guide covers recording your own.
Can I do the Migos AI trend with my own song?
Yes. On MigosAI (migosai.si), choose Your song as the sound, upload the file and select 2 to 15 seconds of it. The lead lip-syncs your audio in the orange booth on Wan 3.0.
Which audio files work?
MP3 and WAV. M4A, the format most phone voice memos save in, is accepted as well. The file can be any length; only the part you select is used.
Does my own song work on Seedance?
No. The Seedance models generate their own vocals from a topic you type, so they cannot use your upload, or the original Hotel Lobby song either. Stay on the default model for your own audio.
Can my friend and I each sing a verse?
Not in a single video: one track, one singer, the person on the left. Make two videos with the photos swapped and join them, as shown in two people, one song.
Does uploading my song cost more?
No. Price depends on model, quality and length only. At 480p on Wan 3.0 it is 3, 4, 5, 6 or 7 credits for 5, 8, 10, 12 or 15 seconds.
Can the video be longer than 15 seconds with my song?
No. With any track, yours or the original, 15 seconds is the cap. The 20 or 30-second videos on Seedance 2.5 only come with an AI rap.
Can I use a famous song I did not make?
Only if you have the right to use that recording. If you want a hit song on your post, the safer route is to add the licensed sound inside TikTok or Instagram when you publish.
What happens to my audio file afterwards?
Uploads, including songs, are deleted 7 days after the render. The finished video stays in My videos until you delete it.
Can I do it on my phone, and how long does it take?
Yes, in any phone browser with no app; the upload button opens the files on your phone. Wan 3.0 usually needs about 5 to 15 minutes, the MP4 has no watermark, and there is no free render: packs begin at 10 credits for $9.99.
Guides
Read next
Two People Lip-Syncing Your Own Song: What Works
Why an uploaded song gives one singer per video, and three ways around it: two swapped videos, an AI rap with Both rap, or one part sung twice
Turn a Voice Memo Into a Hotel Lobby AI Video
From your phone’s recorder to the orange booth: record, save the file, upload it as Your song, frame the best seconds and create
How to Make a Personalized Birthday Rap Video With AI
Three ways to turn two photos into a birthday rap: an AI-written verse, your own recorded wishes, or the original Hotel Lobby duet, plus topics to copy
Bring the track
Give the lead your song
Two photos, one audio file, one orange booth. The sound never changes the price.