Professor of Speech Communication with specialization in Multimodal Embodied Systems
The spoken language is our most important form of communication. A face-to-face conversation is characterised by a complex and carefully orchestrated interplay between different verbal and non-verbal expressions. In addition to the meaning of the words, the conversation includes tone of voice, head movements, glances, facial expressions, articulation and gestures. The subject of speech communication studies the processes that form the basis of human speech communication. It also studies how one can build a system that can manage spoken language, and that encompasses both its verbal and non-verbal components, on a technical level.
Jonas Beskow and his research group are developing and testing models for signals and the phenomena that occur in human speech communication. The objective is to increase the knowledge of how these signals can be used in real systems, such as robots or virtual assistants. The research has led to the development of the social robot, Furhat, which is specifically designed to study the effect of social signals in human-robot interactions. Some examples of applications that are also being studied are assistive devices for increased accessibility through lip reading, and a system for the learning and training of languages. In a recently started project, a robot that will be helpful to the elderly and be able to detect early signs of dementia is being studied.