omnihuman2

AI Singing & Music Performance Video Generator

Have a singer, or a character, perform your track. Upload a photo and a song, and generate a performance clip with expression that follows the music.

Input: Singer photo + song track · Output: MP4, 4–120 seconds

Sample: Music performance · Singer photo + song track. Generated with OmniHuman 2.

How it works

  1. Step 1

    Upload a singer photo or create a performer portrait from a text prompt.

  2. Step 2

    Add your song as a WAV or MP3. Singing uses the same input as speech.

  3. Step 3

    Generate a performance clip and download the MP4.

Why use it

Same input for song and speech

There is no separate singing mode: upload the track and generate.

Expressive performance

Face, head and body movement follow the vocal delivery.

Great for teasers

Create hooks, lyric clips and visualisers for releases. Each video is 4 to 120 seconds long.

Any performer style

Try realistic singers or stylised characters.

Example direction prompts

Optional text you can add in the studio to guide the scene, movement and camera.

Frequently asked questions

Does it really sing?

The video follows the audio you provide. Bring your own song track for singing; a generated speech voice will not sing.

Can I use a copyrighted song?

Only if you own or are licensed to use the recording. You are responsible for the rights to any audio you upload.

How long can the clip be?

Each video is 4 to 120 seconds long. Use a verse or chorus for a clip and edit several together for a longer video.

Is the output a full music video?

It is a performance clip of a single performer. Add scenes, cuts and effects in your video editor for a complete music video.

Can I make an artist look like a real singer?

No impersonation of real people without consent. Use your own artist photos or original characters.

Make your first video

Plans start at $9.99/month. 1 credit = 1 second of video, and failed generations are refunded.

More use cases