Josef is a multidisciplinary scientist with a focus on machine learning for healthcare and audio.
He graduated from TU Munich with a M.Sc. in Electrical Engineering and Information Technology. During this time he enjoyed working part-time as a software developer for helicopter simulators at ESG. His curiosity about decoding the human brain took him to obtaining a Ph.D. in Psychology at TU Darmstadt. He has done several research visits to Seikei University (Tokyo).
While being a postdoc at the University of Cambridge he sped up hearing tests by using Bayesian active-learning techniques and information theory. This allows clinicians to do more tests in a given time, leading to a more precise knowledge about an individual’s hearing. He did further research in basic auditory science.
He won several awards for his work on perceptual models, including a Best Paper and Presentation Award at ICA 2019 and the DIN Award “Young Science” (DIN is the German Organization for Standardization).
His current work is around machine learning for hearing and other healthcare applications. He combines big data techniques and perceptual models for advanced signal processing in hearing aids, and develops active-learning tests to gain personalized perceptual models. In the broader scope of healthcare he wants to apply active learning tests to obtain knowledge about rare diseases and to help doctors decide which tests to do.
In the domain of basic auditory science, he leads experiments on understanding impaired hearing and the perception of noise, and develops models for loudness and speech perception.