How to Choose the Singer in a Raindance AI Video

In a Raindance AI duet, two photos fill two performer roles. One person is selected as the singer and appears in the singing close-ups; the other remains the duet partner. Photo 1 is selected by default, so check the choice before you generate.

Select either person

Upload the first portrait in Photo 1 and the second in Photo 2. The studio marks Photo 1 as the singer when you begin. To choose the other person, use the button beneath Photo 2. You can also choose Photo 1 again with its button. The highlighted selection tells you which person will sing.

The choice follows the photo slot. If you change a portrait, look at the labels and selection again before generating. For example, if the person you want to sing is in Photo 2, select Photo 2 sings even though Photo 1 starts selected. The other person is still included in the duet; singer selection changes who leads the close-ups.

Think of the singer button as assigning the lead role, not removing the other performer. Choose the person whose expression and portrait you want to anchor the close-ups. The partner remains visible as the second person in the seaside duet. If you are unsure, start with the portrait that has the clearest face, then use the button under the other photo if you want to switch roles.

Pick photos that make the roles clear

Use one person per photo, with the face visible and reasonably well lit. A camera-facing portrait gives the studio a clear view of the person, but it does not have to be a formal headshot. A casual photo from a bright room or outdoors can be a useful choice when the face is sharp and unobstructed.

Avoid group shots where it is unclear which face belongs in a slot. Also avoid strong motion blur, deep shadows across the face, large sunglasses, face coverings, or filters that change key features. Similar framing and lighting across both photos can make the pair feel more consistent, though matching backgrounds are not required.

Supported files are JPEG, PNG and WebP. Each image must be at least 300 pixels on each side and 10 MB or smaller. Choose portraits you are allowed to use, and confirm that each photo is in the intended slot before submitting.

No prompt writing is needed

The studio uses the two selected portraits and your singer choice to prepare the duet. You do not need to write, paste or edit a prompt. Set the photos, choose who sings, and continue with Generate video. The output is a 15-second video in 1080p; generation usually takes 1–5 minutes.

If you want the full process—from uploading photos through Google sign-in, a one-time package purchase when needed, and downloading the finished video—read the Raindance AI trend tutorial. You can also open the studio to choose your singer or visit Pricing to review the available video packages.