Flanders Make’s Post

Transcript

Meet our new digital colleague whose job is to process all that information, then provide clear guidance to the operator. As the voice, ears and face of a system, how digital colleague features AI based speech synthesis and generative AI for facial animation. Plus the ability both to detect intent from natural language and provide context aware answers. Both supported by GPT models making it capable of conducting human like conversations more than a robot. Think of this technology as a real colleague observing learning. Knowing and as easy to interact with as any other colleague. The Digital Colleague comprises 3 major AI components. Firstly, the brain. A central knowledge graph organizes all available information to enable cross domain reasoning in a systematic way. This is used to process input from the outside world, understand user intent and formulate suitable instructions to the human operator. Next, the senses and expressions by connecting to all systems. Such as cameras, displays, microphones, speakers and robotic arms, the digital colleague is able to interact with humans and the environment. Finally, we have the underlying nervous system, a modular, scalable open source framework that allows the different senses to interact with the brain, update its understanding of the environment and the question at hand, and provide the most relevant response, whether a clear instruction contextual. Information or a robotic action?