Our research is focused on the design of intelligent signal processing systems that learn their desired behavior through interactive learning with the physical environment. Most projects relate to interactive learning of audio processing algorithms such as personalized speech and music enhancement systems or source separation systems to enhance intelligibility in a cocktail party environment. The common factor to all of our projects is our approach, which is inspired by state-of-the-art work in Bayesian machine learning and computational neuroscience.
I do not maintain a specific list of available projects. I am always interested in discussing both traineeships and graduation projects for strong M.Sc. students with a proven interest and record in machine learning and signal processing. Please also have a look at our starter’s guide to get an impression of our way-of-working.
A few years ago I made a ‘recruitment poster’ and I think it’s still sort-of valid, so here it is: