Google Vids | Gemini Omni Adds AI Video Editing and Personal Avatars

Google Vids is adding Gemini Omni and personal avatars to make AI-assisted video creation more flexible. Creators can generate clips from written instructions and visual references, refine existing footage through conversation, or prepare a digital version of themselves to present a script without recording every scene on camera.


Gemini Omni video editing and personal avatars in Google Vids

{getToc} $title={Table of Contents}

Google Vids turns prompts and references into editable video clips


Gemini Omni allows creators to begin with a natural-language prompt and add images such as a photograph or rough sketch for visual direction. The model combines these inputs to generate a clip that reflects the requested subject, setting, movement, or overall creative idea.


Image references can make the starting point more specific than a text prompt alone. A designer could supply an early composition, product image, character reference, or visual concept before using written instructions to describe how the final scene should move or develop.



Gemini Omni supports step-by-step edits in everyday language


The workflow continues after the first generation. Creators can describe changes such as replacing a background, correcting the lighting, or adding an effect instead of discarding the clip and rebuilding the complete scene from its original prompt.


These conversational edits can also be applied to footage recorded outside Google Vids, including clips captured with a phone. This gives designers a shared editing approach for AI-generated material and traditional video, although each revision should still be reviewed for unexpected changes in objects, faces, movement, and visual continuity.


Personal avatars can present a script without a camera


Personal avatars provide another way to appear in a Google Vids project. After uploading a selfie and a short voice recording, the user can type the message they want to communicate and have the digital avatar deliver it with their appearance and voice.


This can support recurring updates, introductions, instructional clips, and personalized messages when recording a new performance would interrupt the workflow. Each avatar is linked to the user's Google Account and restricted to the account holder's likeness, limiting the feature to creating a digital representation of oneself.


Availability, identity limits, and SynthID transparency


Gemini Omni and personal avatars are available in Google Vids for Google AI Pro and Ultra subscribers and Google Workspace business customers. Personal avatar access is more restricted, as users must be at least 18 years old and located in a supported region.


Every AI-generated clip includes an invisible SynthID digital watermark. The watermark provides a way to identify the use of generative AI while allowing creators to share the finished video without adding a visible label over the composition.


IMPORTANT: Personal avatars are available only to eligible users aged 18 or older in supported regions. They are connected to the user's Google Account and cannot be created from another person's likeness.{alertWarning}

Daisuki's Take: What This Means for Designers


Gemini Omni makes Google Vids more useful as an iterative creative workspace. The ability to combine prompts with visual references and refine the result gradually can preserve a promising composition while correcting individual problems that would otherwise require another complete generation.


Personal avatars may also reduce the production work behind internal presentations, tutorials, and recurring video messages. They do not replace the expressive control of a real performance, but they provide a practical option when the information matters more than staging and recording a new scene.


We would still inspect every clip before publishing it. Prompt-based edits can alter details beyond the requested area, and an avatar that represents a real person requires careful review of its voice, appearance, script, and context before it becomes part of client work or public communication.



Sources and Recommended Links