Pierre Divenyi : Fishing a Sound Pattern Out of the Mud - Some Experiments In Auditory Scene Analysis

Wed, 02/20/2013 - 5:15pm - 7:00pm
CCRMA Classroom, The Knoll 2nd floor, Rm 217
Event Type: 
 When the objective is to recognize a target pattern in a setting where other streams are concurrently present, the process can be challenging, especially when the target and the distractor streams share characteristics along one or more dimensions. In a series of experiments, we examined this phenomenon – failure of auditory scene analysis also called informational masking – in reduced settings.
The target pattern is a harmonic complex of 500 ms or shorter having a given fundamental frequency, presented simultaneously with a single distractor that is also a harmonic complex but of a different fundamental frequency. In one set of experiments, the relevant dimension of target and the distractor was their particular amplitude modulation pattern, whereas in another set of experiments it was two different patterns that modulated the center frequencies of two broad band-pass filters following the target and the distractor harmonic complexes, respectively. Results show some predicted effects, such as the increased disruption of target recognition when the fundamental frequencies of the two streams are close. However, there were a few surprises as well: actual overlap of the target and distractor frequency ranges, i.e., energetic masking, accounted for very little disruption; salience of the target was always higher when its fundamental frequency was higher than that of the distractor (which, actually, would surprise only firm believers of masking profiles, not musicians); FM patterns, except those with small FM excursions, were easy to recognize even in very low target-to-distractor level ratios; AM patterns were recognizable even at very rapid, i.e., high AM-frequency, rates; the degree of AM target-distractor asynchrony had an nonmonotonic effect on recognition – disruptive at very brief and actually helpful at moderately long asymmetries; performance of elderly listeners was consistently poorer than that of the young. Implications of the findings to the “cocktail-party effect” and to music listening will be discussed.


Pierre Divenyi started his career as a pianist, giving recitals in Europe and the US. As a graduate student at the University of Washington, his interests turned toward science and obtained his doctorate in systematic musicology, writing a thesis on the perception of rhythm in micro-melodies. That work led him to studies on the psychoacoustics of tone sequences, time intervals, and auditory localization. He worked as a researcher first at Central Institute for the Deaf in St. Louis and then at the Martinez (CA) VA Medical Center’s Speech and Hearing Research Laboratory, which he directed from 1979 to 2012. His research over the past two decades has been focused on auditory scene analysis and, in particular, on the auditory processes underlying the separation of speech from background noise – the so-called “Cocktail-party effect” – and its dysfunction in aging. In search of a solution to this problem, he organized several international multidisciplinary meetings that brought together psychoacousticians, auditory neuroscientists and computer scientists. Over the last decade, he edited two books on empirical and computational research on speech separation and was recipient of several collaborative interdisciplinary grants to study the issue. He has been visiting professor and visiting scientist at several universities in the US, Canada, France, and Germany, and has been frequently invited to lecture both in the US and abroad. He joined CCRMA as a Consulting Professor in early 2012.
Open to the Public
Syndicate content