Laboratory for digital signal processing

In the laboratory we are developing a new concept of conversational language understanding (CLU) as a new unique approach that develops the idea that verbal and nonverbal conversational cues are complementary, and equally important in conversation. Key research activities are focused on understanding and modeling conversational language in human-to-human interactions. In the field of natural language processing and understanding of spoken language, gesture and non-verbal communication, as well as the ability to express information not only through words, begin to play an important role in human-machine interaction. It has become clear that the integration of such signals is a key direction in identifying a more personal aspect of user inputs and how device responses can be presented to people. In direct interaction, non-verbal signals transmitted along with spoken content (or even in the absence of it) are crucial for establishing cohesion in the discourse. In the laboratory, we are developing new fusion-based models as well as artificial intelligence (AI)-based algorithms that will be able to generate in-depth understanding in cognitive interplay with communicative intent as a central core in human interaction. The algorithms are based on deep learning techniques.