Josh McDermott, PhD
Fred & Carole Middleton Career Development Assistant Professor
Department of Brain and Cognitive Sciences
Massachusetts Institute of Technology
PI, Laboratory for Computational Audition
jhm - AT - mit - DOT - edu
I study how people hear. My lab investigates all things auditory, operating at the intersection of psychology, neuroscience, and engineering. Our longterm goals are to understand how humans derive information from sound, to improve treatments for those whose hearing is impaired, and to enable the design of machine systems that mirror human abilities to interpret sound.
My CV (with full list of papers and links to pdfs)
Synopsis of Research Interests:
Amazing Success of Biological Auditory Systems
Humans routinely perform tasks with sound that remain impossible for even the most powerful and sophisticated machine hearing systems. Following a conversation on a noisy city street, recognizing the sound of keys in a door, learning the sound of a new word - we do such things every day without a second thought. Their difficulty is revealed when we attempt to build machines that replicate our abilities. Understanding how we hear in these situations involves confronting the most challenging computational problems in audition.
I try to do experiments in humans that reveal how we succeed in situations where machine systems fail, and to use results in machine hearing to motivate new experimental work. Recent work in this vein has targeted sound segregation, sound recognition, and the perception of reverberation.
I spend a lot of time studying what naturally occurring sounds are made of, as this holds many clues to how we hear them. Developing good models of natural sounds also allows us to generate novel naturalistic sounds, which have many uses in experiments.
I have long-standing interests in the science of music. I continue to think a lot about what makes music pleasurable, why some things sound good and others do not, and why we have music to begin with. These are big questions, but the right experiments have potential to provide insight. Music also provides great examples of many interesting phenomena in hearing, and as such is a constant source of inspiration for basic hearing research.
I started college intending to study physics and math, but was soon seduced by the mysteries of the brain – in particular, its stunning ability to solve ill-posed perceptual problems. My early training was in vision. After finishing a BA in Brain and Cognitive Science at Harvard, I headed to London to study at the newly formed Gatsby Unit, where I completed an MPhil in Computational Neuroscience. I then returned to the US for a PhD in Brain and Cognitive Science from MIT. During grad school I got interested in sound and hearing, and eventually transitioned into auditory research, with postdoctoral training in psychoacoustics at the University of Minnesota and in computational neuroscience at NYU. In 2013 I joined the Department of Brain and Cognitive Sciences at MIT as an Assistant Professor, where I currently lead the Laboratory for Computational Audition.
A not-so-recent interview
A page with most of my vision papers.
A tutorial of all my motion demos.