How to make a Hotel Lobby AI video.

Prepare two separate portraits.

To create a Hotel Lobby AI video, select one photo of each adult. Both people must agree to having their likeness used in the AI video. A shared group picture is harder to use; each reference should contain one clear face.

• Use a sharp, well-lit image with the eyes, nose and mouth visible.
• Choose a front-facing or gently angled face. Avoid a full side profile.
• Avoid sunglasses, hands over the face, heavy filters and motion blur.
• Use JPEG, PNG or WebP, up to 10 MB per portrait.

Put the first person in Left performer and the second in Right performer. The slots determine their roles in the source clip. Use Swap sides before generating if they are in the wrong order.

Watch the AI demonstration.

Play the 10-second fictional-duo demonstration with sound. Review the two performer roles and the example movement. The demonstration is separate from the selected 5-, 12- or 30-second generation length. The current studio uses a fixed performance template; you are choosing the cast for that footage. The Migos Hotel Lobby AI page explains how this source determines the finished video.

Every prepared segment begins at 0:00 of the original video. Choosing a longer length includes more of the same performance from the beginning. It does not select a later section or create a new introduction.

You need permission to use the source footage and audio, as well as both portraits. Generating an edit does not grant music rights or permission to publish someone’s likeness. Read the usage terms before proceeding.

Choose length, resolution and frame.

The studio offers the first 5, 12 or approximately 30 seconds, with 480p or 720p output. You can keep the original 16:9 frame or crop to 1:1 or 9:16 after generation. Vertical and square crops cut the sides of the two-person shot; they do not restage the performance. The exact prepared duration is shown beside the player.

Portrait photos do not by themselves change the finished video into a vertical format. Choose 9:16 if you want a vertical file.

Sign in before generating. The studio shows a credit quote for the selected settings. If you need more credits, choose a one-time pack on credits and pricing.

Review before you generate.

Confirm both adults’ consent and your permission to use the source material, then open Review video. This dialog lets you check the two portraits, left and right assignment, source segment, resolution and quoted credits.

Opening the dialog does not generate a video. The final Generate button reserves the quoted credits before submitting one video generation request. A completed video settles the reservation; a definitive failure returns it once. An uncertain submission keeps credits reserved until the existing request is resolved. Stay on the page while the request is being saved; the result page will open afterward.

If the connection times out, open your last request before trying again. A missing response does not prove that the provider rejected the generation.

Compare, listen and download.

The result page shows the AI demonstration and your generated video separately. The application restores the original segment’s audio after the provider finishes. A completed result becomes available for MP4 download after media checks. Watch the completed example and the AI demonstration to see what to compare.

• Check the opening scene and watch for unwanted introductions.
• Check the left and right faces throughout the clip.
• Watch hands, expressions, clothing and background for visual changes.
• Listen for the original track, then check mouth timing against it.
• Review the duration, framing and any timing adjustment noted on the page.

AI edits can vary in likeness and movement. Original audio restoration does not guarantee exact lip sync. Keep the result private until you are satisfied and have permission to share it.