Binaural Processing of Multiple Sound Sources

Project: Research project

Description

In the real world a listener is confronted with many sound sources, and often several of those sources produce sounds at about the same time. Understanding how those sound sources are perceptually segregated has been a challenge for hearing science for more than one hundred years. This proposal deals with the ability of the auditory system (most probably, the binaural auditory system) to extract information about the interaural time differences and interaural level differences of individual sound sources when more than one source produces simultaneous sounds as occurs in the real world (i.e., multiple-sound source processing in the azimuth plane). The challenge facing the auditory system with multiple sound sources is that the sounds interact often obscuring key features like interaural differences. The objectives of this proposal are to: 1) Describe the performance of human listeners in localizing sound sources in a wide variety of multiple sound source situations. 2) Explore a representation of the interaural information human listeners might use in these multiple sound source conditions in order to develop a model of processing interaural differences that predicts localization performance related to individual sound sources when several sound sources produce simultaneous sounds. Being able to extract interaural cues related to the individual sound sources in such multisource environments has practical relevance in at least three domains: 1) Localization of sound sources in a real world of multiple sound sources. 2) Detecting-discriminating-recognizing one sound (target) in the presence of other competing sounds (maskers) when the target and maskers are spatially separated (i.e., spatial release from masking in the real world), and 3) The overall real-world problem of perceptually segregating several sound sources, especially when the sounds occur simultaneously. This proposal will focus on localization of individual sound sources when more than one source produces simultaneous sound.
StatusFinished
Effective start/end date6/1/127/14/16

Funding

  • DOD-USAF-AFRL: Air Force Office of Scientific Research (AFOSR): $805,795.00

Fingerprint

acoustics
proposals
cues
hearing
masking
azimuth