Gemini launches feature to create AI avatars with your own face and voice

Gemini has launched a feature that lets users create photos and videos with artificial intelligence (AI) using their own face and voice. Through the Gemini app's new 'Personal Avatar' feature, users can build a digital replica modelled on their own face and voice.

The avatar can then be used to create different kinds of photos and videos. As a result, users will not need to record a new video of themselves every time; the avatar can appear on screen in their place.

Another notable aspect of the new feature is that users can share their avatar with others. By sending an invitation to the Gmail address of a friend, family member or colleague, the owner can grant permission to use the avatar. Those who accept the invitation can use the avatar owner's face and voice in their own AI-generated photos and videos.

However, the avatar's owner will control who can use it and for how long. Gemini's settings include a separate arrangement for managing these permissions; access can be granted for a set period and revoked at any time if needed.

Once permission is revoked, the person concerned can no longer create new photos or videos with that avatar. But photos and videos created earlier, while the permission was in place, will remain intact; withdrawing permission does not delete previous content.

According to Google, creating a personal avatar first requires taking a photo of one's face as well as recording one's voice. If the process is started from a computer, Gemini will display a QR code. After scanning the code with a smartphone or tablet, the voice has to be recorded on the phone. Once the avatar is created through Gemini, it can be used later to make new photos and videos.

However, a Google AI plan will be required to create videos using a personal avatar.