VEO Omni: AI Video Where Characters Speak on Screen

VEO Omni is Google's Gemini-based video model built for one thing: a character on screen speaks, and the sound and lip movement are produced together in a single pass. On MarketVideo you run VEO Omni 1.1 from the Create video page.

Make a video with speech

Ready in minutes ยท Free templates every day

Older workflows took two steps, rendering the clip and then laying a lip-sync pass over it, and the seam often showed as mouths drifting off the words. With Omni that second step simply is not there.

  • Speech and picture in one render

    Inside VEO Omni the image and the audio are built at the same time instead of being glued together afterwards, which is what the Omni name points to. A character says the line, and the mouth shapes follow the words within the same render, so there is no separate sync stage to fix later.

  • Writing a line for VEO Omni 1.1

    Choose VEO Omni 1.1 among the models offered on Create video, then lay out the shot: who stands in the frame, where they are and what they say. Write the dialogue word for word and keep it to a sentence or two. When a particular face should appear, attach a picture of you or of a person who gave consent.

  • VEO Omni versus Veo 3.1

    Veo 3.1 is the choice for the overall sound of a scene: ambience, background noise, the mood of a place. Omni is the pick when a person on screen has to deliver words with matching lips. The templates in the grid were built on Veo, so they show finished Veo scenes rather than Omni runs.

  • Limits worth knowing

    Each clip is one short scene, and a word can come out blurred, so a second take is sometimes needed. Rendering is a matter of minutes, and queues can add more. Celebrities and public figures are not allowed; children appear only in your own family videos.

Ready-made templates

FAQ

Are Gemini Omni and VEO Omni the same thing?

They name one line. VEO Omni is built on Google's Gemini family, the same family behind Nano Banana for images, so a search for Gemini Omni leads to this model.

Do I need a separate lip-sync step?

No. Speech and mouth movement come out of one render. LTX-2.3 LipSync serves a different job: matching a portrait to a recording you already have.

Does MarketVideo add a voice of its own?

There is no standalone voice tool on the site. Whatever the character says is produced by the video model itself as part of the clip, based on the line you wrote.

Can I try VEO Omni 1.1 for free?

Sign-up costs nothing, and a newly verified account gets 5 stars to spend. Check the star price shown for VEO Omni 1.1 before running; if your balance falls short, Pricing has plans and packs.

Related