Artificial intelligence models can naturally make virtual avatar gestures to match spoken words, potentially allowing AI-generated newsreaders and influencers to move more realistically while speaking. There is.
When humans speak, we use gestures to convey meaning. But when video game characters or digital avatars try to perform similar actions, they often make generic moves, regardless of what they're actually saying. To make virtual people's gestures more realistic, researchers first needed to teach the AI model the relationship between voice, body language, and emotion…
Source: www.newscientist.com