Major Google Vids update: Create custom "AI Avatars" from selfies, powered by Gemini Omni for text-to-video creation
Google introduces new capabilities to Google Vids, allowing users to create digital avatars for their videos without manual filming, powered by the Gemini Omni model for easier AI-driven video production.
Google has announced a major update for Google Vids, its Workspace video creation tool, featuring two key additions: personal avatars and the integration of the advanced Gemini Omni AI model.
The new personal avatars feature allows users to create digital avatars with their own likeness and voice simply by uploading a selfie and a voice sample. This enables users to be the "protagonist" of their videos without needing a camera setup or professional filming, making it highly efficient for presentations and corporate communications.
Additionally, Google Vids is now powered by Gemini Omni, a high-performance AI model that offers more flexible video creation and editing. Users can generate video clips from text prompts or reference images and edit mobile-recorded footage directly—such as changing backgrounds, adjusting lighting, or adding effects. It also supports step-by-step editing, allowing users to fine-tune content without starting from scratch.
To ensure safety and transparency, Google has implemented strict oversight: personal avatars are tied exclusively to the creator’s Google account, access is restricted to users 18+ in select regions, and all generated videos will include SynthID, an invisible digital watermark for traceability. These features are available to Google AI Pro/Ultra subscribers and Google Workspace business customers.
This update streamlines the creation of professional, speaker-led videos, removing barriers for those without filming equipment or the comfort of appearing on camera, which significantly lowers the barrier to entry for professional presentations, training, and marketing content.