Deep Brain AI (CEO Jang Se-young), which focuses on artificial intelligence (AI), announced on the twenty third that it should introduce 3D virtual human B2B technology that permits real-time conversation in addition to emotional expression and three-dimensional motion.

3D virtual human is a 3D rendering-based virtual character that may set clothes, hair, language, and tone from various angles, unlike existing live-action AI humans optimized for frontal speech. It’s explained that there are control points on the character’s face and body, so it is feasible to create a three-dimensional character and space much like reality, comparable to sitting in a chair.

It’s a 3D virtual human based on Unreal Engine, enabling real-time conversation and not using a separate video compositing process. It is claimed that it is provided with a real-time response that expresses emotions or gestures that match the user’s facial features and eliminates the time required for rendering before outputting speech, enabling natural motion.

As well as, after preparing various speech shapes, he emphasized that voice data will be subdivided, and natural mouth shapes and pronunciations in addition to emotional expressions and movements are possible. It was also cited as a strength that it may possibly be placed in various spaces comparable to VR and metaverse.

▲ Asian male model (Yuri) ▲ Western male model (Peter) ▲ Western female model (Sophia) ▲ Black female model (Amber).

Jang Se-young, CEO of Deep Brain AI, said, “Following the present 2D real-life AI human, we’ve got expanded the range of decisions in order that it may possibly be utilized in various fields by implementing a highly complete 3D virtual human.” “Revolutionary AI through constant research and AI-related technology development We are going to take the lead in providing recent business opportunities with human services.”

Reporter Juyoung Lee juyoung09@aitimes.com