At CES 2022, Deep Brain presented photorealistic avatars that companies are invited to use to communicate with customers. Avatars are generated based on several hundred hours of videos of real people.
DeepBrain avatars are able to inform and guide users in thousands of possible scenarios and interactions in real time.
The Deep Brain model is trained to reproduce individual features of speech, accent and gestures on several hundred hours of recorded video of real people – prototypes of each avatar. The avatar pronounces the script, imitating these features and copying the appearance of people. The model includes speech synthesis, video synthesis, and real-time natural language processing.
DeepBrain avatars are already used by companies such as 7-Eleven, KB, LG HV and Roche. The technology can be implemented taking into account the company’s industry. Avatars-announcers, presenters, bankers, teachers, presenters, concierges, doctors and lawyers were presented at CES 2022.