Synthesia, a British startup, has launched “Expressive Avatars”, a new generation of AI models designed for advanced emotional expression in video creation. These digital actors, trained on real actor footage, respond to text prompts with nuanced emotional displays. Their debut marks a shift from static digital renders to dynamic performers capable of unique, repeatable expressions.
Watch the launch event for the expressive AI avatars right here:
While showcasing promising potential, these avatars still exhibit exaggerated emotional traits and limited physical movement, hinting at evolving roles in training and customer service videos. Despite their advancement, concerns about their use in spreading disinformation persist, prompting Synthesia to strengthen platform regulations and content moderation.
So…:
How can you utilize AI-generated video elements where the human component is now even more lifelike?
Source & Picture: