Speech perception, language representation and processing in artificial neural networks and the brain

Auditory (phantom) perceptions, hearing disorders, tinnitus, hyperacusis, pathological speech

Deep artificial neural networks for modeling brain function

Biologically plausible computational models of cognition

Analysis and visualization of high-dimensional, massively multi-variate and multi-modal neuroimaging data (M/EEG, fMRI, invasive cortical recordings) acquired during speech perception and prodcution based on existing data science methods (machine learning, classical and Bayesian statistics) and self-developed methods

Application of neuroscience methods and development of novel methods for the analysis, reverse-engineering, visualization and interpretation of neural networks (black box problem, explainable AI)

Investigation of structural and dynamical properties of recurrent neural networks (RNNs) with methods and concepts from theoretical physics (information theory, dynamical systems and chaos theory, random matrix theory, theory of complex systems, network and graph theory)

Transfer of information processing principles from neuroscience to machine learning and AI systems (neuroscience-inspired AI)

Neuroscience of sleep and analysis of sleep stages