YouTube is now letting users create an AI avatar that looks and sounds like them for use in Shorts. This feature was teased earlier this year.
The creation process is available in the main YouTube app and YouTube Create. It involves capturing a “live selfie” by recording your face and voice (by reading a few prompts).
This will give you a photorealistic avatar that can be placed in your YouTube Shorts. Each prompt-based generation can be up to 8 seconds long, though you can create multiple clips back-to-back. You only have to do the setup process once, but can retake at any time to update your appearance.
Prompt and output
The feature is a continuation of Google Veo models in YouTube Shorts. Something similar has been possible with the ingredients-to-video capability, which involves uploading a picture, since last year. However, the voice aspect is entirely new.
YouTube tells 9to5Google that “the avatar feature gives users an easier way to include themselves safely and securely in videos.”
On the safety front, the selfie video and voice will only be used for avatar creation, and no one else can use the avatar to create original Shorts. You can delete your avatar at any time (with YouTube automatically deleting after 3 years of no use), but keep in mind that existing videos with your avatar will remain until the actual clip is deleted.
All avatar videos will feature watermarks and digital labels, like SynthID and C2PA, as well as disclosures about AI-generated content.
Once rolled out to you, open the Create ‘+’ in the YouTube app and tap the Gemini spark in the corner. Select “Create video” in the top-left and look for “Make a video with my avatar” to enter your prompt. It’s also available in the Remix menu > Reimagine > Add me to this scene.
This feature is rolling out starting today globally (outside of Europe) for users 18+, with full availability in the coming days. You must be the owner of an existing YouTube channel to create an avatar.
FTC: We use income earning auto affiliate links. More.
Comments