Cool or creepy? Microsoft's VASA-1 is a new AI model that turns photos into 'talking faces'

Microsoft's VASA-1 is an advanced AI developed by Microsoft Research that transforms a single portrait photo and an audio file into a hyper-realistic talking face video. This AI model showcases impressive lip-sync capabilities and realistic facial and head movements. Although it is currently only a research preview and not available for public use, VASA-1 stands out for its ability to process various facial orientations and incorporate emotional nuances into animated characters, suggesting potential future applications in gaming, virtual avatars, and AI-driven media production.

Opens on Tom's Guide