Google Vids creates a digital avatar of the user using AI

Google has announced a major update for its Google Vids platform. The service now allows users not only to use ready-made templates but also to create personal digital avatars that mirror their appearance and voice. This technology is expected to fundamentally simplify the video creation process and usher in a new era for corporate communication and education. This is reported by Techcrunch.com reports .

To use the new feature, the user simply needs to upload a selfie and a voice sample to the system. The AI then creates a digital twin that speaks the provided text just like a real person. According to ixbt.com, this update transforms Google Vids from a simple presentation tool into a full-fledged video editing platform.

Gemini Omni integration and editing capabilities

Google Vids is now powered by the Gemini Omni multimodal AI model. This allows users to create complex videos using text prompts and uploaded images as references. The system can not only generate new footage but also replace backgrounds in phone-recorded videos, adjust lighting, or add various visual effects.

One of the most important innovations is the step-by-step editing function. Previously, if there was a minor error in an AI-generated video, one had to start from scratch. Now, with the help of Gemini Omni, it is possible to make precise changes to any part of the video, edit individual frames, or replace elements.

Security and usage restrictions

Google is taking strict measures to prevent the malicious use of its digital avatars. Each created avatar is linked to the user's personal Google account, and videos are embedded with invisible watermarks using SynthID technology. This helps identify content as AI-generated.

For now, the ability to create personal avatars is said to be available only to users over 18 in specific regions. Google primarily intends this tool for Google Workspace users, i.e., business representatives. This technology will significantly save time and money when preparing internal company news, employee training, and tutorials.

With this step, Google is entering into serious competition with AI-based video services like HeyGen, Synthesia, and D-ID. While OpenAI is delaying the public release of its Sora model, Google has already begun integrating a finished product into its ecosystem.