How voice familiarity facilitates intelligibility of degraded and masked speech

点击翻译按钮获取中文摘要

Following and understanding one particular conversational partner, despite interference from other sources, is a feat most of us accomplish every day. However, many processing operations are required so that, when hearing a complex auditory signal consisting of many different sources of sound (and their echoes), one source in particular (i.e., a voice) can be identified, tracked, and understood. My trainees and I study how experience and prior knowledge feed into voice listening and comprehension, particularly in noisy, everyday, conditions. Knowledge-guided processes (e.g., the use of meaningful context to perceptually fill in' masked or degraded speech) can dramatically improve perceptual organization and speech perception, but often require attention, which is probably why communicating in noisy social situations is effortful and often frustrating. One underexplored knowledge-based factor that may enhance comprehension while actually reducing listening effort is familiarity with a talker's voice. Johnsrude et al (2013) demonstrated that people can exploit knowledge of a highly familiar voice (their spouse's) not only to track it better in the presence of an interfering stranger's voice, but also to ignore it so as to comprehend a stranger's voice more effectively. This preliminary study indicates that cues derived from experience with particular voices are potentially highly significant facilitators of speech comprehension in naturalistic situations; but almost nothing is known about them. The research proposed here will begin to fill this gap, through four aims:1) to examine how much experience, and what kind, is required for voices to become familiar so as to benefit intelligibility; 2) to identify the acoustic characteristics of voices that are most important for enabling familiar-voice benefit; 3) to examine whether voice familiarity reduces listening effort; 4) to compare the neural (and cognitive) mechanisms supporting perception of speech in familiar and unfamiliar voices.The goal of my research program is to develop a more detailed theoretical account of how familiar voices contribute to comprehension of degraded and masked speech. This work is transformational both for our understanding of the cognitive processes underlying speech comprehension, and for tackling the problem of impaired real-world speech comprehension. It is a necessary first step to: developing hearing prostheses that preserve essential cues that enhance intelligibility of a voice; enabling us to improve listening environments in schools and public places so familiar-voice information can be exploited; and enabling older people to take more complete advantage of familiar-voice information in challenging listening environments.