Torsten Dau on Using data-driven auditory profiling to scene-aware signal processing for super-human hearing
Date:
Thu, 11/29/2018 - 10:00am - 11:30am
Location:
Seminar Room at the Knoll
Event Type:
Hearing Seminar So, Torsten Dau **Thursday at 10AM****
And Pragati Rao, who was going to speak on Friday, will be leading a discussion of her musical EEG work in January.
I’m really happy that Torsten Dau will be here at CCRMA on Thursday morning to talk about the work he is doing on auditory profiling and scene-aware signal processing to enable all of us to have better hearing. From the day we are born, our hearing degrades (although I like to think I am smarter now than I was then. :-) Hearing devices take many forms, from phones to dedicated devices, and all of them manipulate the auditory world, allowing us to hear what we otherwise can’t hear. Or perhaps making it easier to hear.
Torsten is a really smart auditory scientist from the Danish Technical University. He has made fundamental contributions to the perception of speech, modulation, masking, and fine-time structure, as well as computational models of all the above. He consults with a number of hearing aid manufacturers, and he is a really nice guy. I’m sure we’ll have a wonderful discussion.
Who: Torsten Dau (Technical University of Denmark)
What: Using data-driven auditory profiling to scene-aware signal processing for super-human hearing
When: >>>Thursday November 29, at 10AM. <<< New time and date!
Where: CCRMA Seminar Room (Top floor of the Knoll at Stanford)
Why: We need new paradigms for hearing assistance, for all
I apologize for the short notice. I hope all of you can make it. Torsten is wonderful. This coming Thursday morning at CCRMA. Don’t miss it!!!!
Bring your favorite ears and we’ll talk about how to improve them.
- Malcolm
Using data-driven auditory profiling to scene-aware signal processing for super-human hearing
Torsten Dau - DTU - Technical University of Denmark
Despite advances in acoustic technology, modern hearing aids have yet to solve the fundamental problem of restoring hearing in everyday sound environments. Finding the best compensation strategy for an individual person represents a major challenge since the consequences of a typical hearing loss are much more complex than just a reduced sensitivity to sound as reflected in the pure-tone audiogram. In fact, characterizing the “auditory profile”, which would require many more measurements than are currently conducted in audiology clinics, seems essential for optimizing the amplification strategy in hearing-aid fitting for each individual. Furthermore, while normal-hearing listeners are able to focus attention on one particular sound source and ignore others, this ability is reduced in listeners with hearing impairment. The crucial problem of modern hearing aids is that they do not know which acoustical source an impaired listener would like to hear. To solve this problem, hearing aids need to evolve from sound processors to 'brain processors' collecting information from the listener to selectively amplify only the sounds that the listener is trying to focus on. Such a revolution requires significant breakthroughs in terms of our fundamental understanding of hearing. Specifically, we need models that bridge the fundamental gap between sound processing in the inner ear and processing in the brain. The ability of the auditory system to extract meaningful 'auditory objects', like speech or music, from a mixture of sound waves arriving at the ear involves multiple stages of processing. The goal of modern hearing research and technology is to develop functional models of hearing that integrate these levels of processing to investigate how the 'listening brain' actively modulates sound processing to serve behavioral goals.
Torsten Dau was born in Hannover, Germany. He received the diploma degree in Physics at the University of Göttingen (Germany) in 1992, and the Dr. rer. nat. degree in Physics at the University of Oldenburg (Germany) in 1996. From 1997-2002, Dr. Dau was assistant professor (C1) at the Physics Department in Oldenburg. From 1999-2000, he was visiting scientist at the Department of Biomedical Engineering (BME) at Boston University and the Research Laboratory of Electronics (RLE) at Massachussets Institute of Technology (MIT), USA. In 2003, he received the habilitation degree (Dr. rer. nat. habil.) in Applied Physics at the University of Oldenburg. Since 2003, he is professor for Acoustics and Audiology, head of the Center for Applied Hearing Research (CAHR) and head of the group “Hearing Systems, Speech and Communication” at the Technical University of Denmark (DTU). His main research interests are in the fields of auditory signal processing and perception, technical and clinical audiology, auditory neuroscience, neural modelling, and technical applications of auditory models in hearing technology. Dr. Dau has published various journals and conference papers, edited three books and organised several international conferences. He is fellow of the Acoustical Society of America, has been guest editor for several international journals and a member of several acoustical and audiological societies.
FREE
Open to the Public