How to Make an Alors on Danse AI Video

Making an Alors on Danse AI video is mostly a photo-selection problem. The template already supplies the white studio, keyboard, laptop, camera cuts, and performance rhythm. Your two uploads control the identities.

Prepare two portraits

Use one clear, front-facing photo for the workstation host and one for the visitor. Each upload should contain one person. Keep the face large enough to read and avoid heavy beauty filters, sunglasses, and extreme crops.

The first photo controls the person at the keyboard and laptop. The second controls the person standing beside the desk. This is the role mapping that prevents cross-swaps during close-ups.

Choose the output frame

The source is a landscape studio setup, but the generator can prepare different output canvases:

• 16:9 for embedded posts and YouTube
• 9:16 for TikTok, Reels, and Shorts
• 1:1 for mixed feeds
• Auto when you want the model to match the source

The action stays the same. Only the canvas changes.

Pick the resolution

480P costs 50 credits and is useful for testing a pairing. 720P Standard costs 100 credits and is the practical default. 1080P Pro costs 200 credits and is the higher-detail option.

The form shows the price before submission. Provider-side failures also return unused credits.

Generate and inspect

Submit the task and follow the progress panel. The result appears when the provider returns the rendered video. Check both identities across the wide shots and the close-ups, then download the version you want to keep.

Avoid the usual mistakes

Do not put a group photo in either slot. Do not reverse the workstation and visitor roles. Do not upload the same person twice unless you specifically want both roles to share one face. Do not expect the model to invent new dialogue or a new scene. The product works because the template is fixed.

If you want to use another model, start with the prompt library. If you want the shortest route, use the generator and upload two photos.