Vismuse

Mabel Matiz Umrumdışı AI Video Generator

Create a Mabel Matiz Umrumdışı AI piano duet from two photos: a singer on the left, a pianist on the right and matching source sound in a 14-second vertical clip.

Create my video

Pick an effect

All 136 effects
0 / 3000

Paid credits required. Check the cost before generating.

Reference preview
Demo preview. Actual results depend on your images and instructions.

What happens in this video?

The Umrumdışı trend turns a music-room performance into a character or pet duet: one subject sings with their chin lifted while the other plays the keyboard. Also searched as Umrum Dışı, Umrumdisi and Yan Sen Derdine, this scene depends on two distinct roles, expressive mouth movement and visible key presses. The prepared performance guides the timing; precise lip-sync and facial continuity can vary.

Demo preview. Actual results depend on your images and instructions.

Your photos and the prepared scene

Photo 1 is the singer on the left. Photo 2 is the pianist on the right, with a plain wristwatch as a scene detail. Each uploaded subject replaces the complete reference performer, not just their face. Their own species, markings and photographed clothes remain. The classroom, guitars, keyboard, gestures and camera timing come from the prepared scene; no video upload is needed.

Create my video

  1. 01

    Add your photos

    Add two separate photos in role order: singer first, pianist second. Show one face and the upper body or whole body in each image, with clear hands or paws where possible. Pets retain their own species and markings. Avoid group shots, faces hidden by accessories and heavy filters.

  2. 02

    Personalize the scene

    Photo 1 is the singer on the left. Photo 2 is the pianist on the right, with a plain wristwatch as a scene detail. Each uploaded subject replaces the complete reference performer, not just their face. Their own species, markings and photographed clothes remain. The classroom, guitars, keyboard, gestures and camera timing come from the prepared scene; no video upload is needed.

  3. 03

    Generate and download

    Check the credit cost, generate, then inspect faces, hands and transitions before downloading. New generations and revisions use credits.

Make it your way

The template is built in. Use the ready-made scene, or add a short instruction to make it yours.

Ready to try it? Upload and generate.

Add two separate photos in role order: singer first, pianist second. Show one face and the upper body or whole body in each image, with clear hands or paws where possible. Pets retain their own species and markings. Avoid group shots, faces hidden by accessories and heavy filters.

No prompt needed. Leave the optional instructions blank to use the included template.

Make your video

Have something specific in mind?

Describe the clothing, lighting or setting you want in the optional instructions. A short, specific request is enough.

For example
Keep the tabby's markings and the dog's blue collar. Let the singer tilt the head up naturally and the pianist press separate keys in a warm music classroom.

2K is available on eligible plans and uses more credits than 768p.

Questions about this effect

Which photos should I upload?

Add two separate photos in role order: singer first, pianist second. Show one face and the upper body or whole body in each image, with clear hands or paws where possible. Pets retain their own species and markings. Avoid group shots, faces hidden by accessories and heavy filters.

Do I need to upload a reference video?

Photo 1 is the singer on the left. Photo 2 is the pianist on the right, with a plain wristwatch as a scene detail. Each uploaded subject replaces the complete reference performer, not just their face. Their own species, markings and photographed clothes remain. The classroom, guitars, keyboard, gestures and camera timing come from the prepared scene; no video upload is needed.

Does the video include sound?

With sound enabled, the matching audio is extracted from the prepared video and added after silent visual generation. A silent request leaves it out. Exact beat alignment can vary.

How long is the video?

One 14-second 9:16 MP4. Check the face, body movement and scene continuity before downloading. New generations and revisions use credits.

Can I change the outfit or background?

Yes. Add a short instruction about clothing, lighting or the setting. Keep the assigned photo roles for the selected scene.

Create my video

Add two separate photos in role order: singer first, pianist second. Show one face and the upper body or whole body in each image, with clear hands or paws where possible. Pets retain their own species and markings. Avoid group shots, faces hidden by accessories and heavy filters.

Create my video

Explore more video trends

On The Radar AI Video GeneratorTiny Desk AINEON Dance AI