Electronic models to simulate the effect of sensory distortions on speech perception by the deaf

Abstract

Electronic models that process signals to simulate omissions and distortions in perception by the deaf are useful analytical tools. These models can help in determining the degree to which the represented loss and/or distortion of acoustic speech cues, isolated from' any central pathology, affects speech intelligibility. They can also help in evaluating the potential effectiveness of different modes of compensatory signal processing. Three examples of such models, designed on the basis of measured characteristics of the deaf subjects' residual hearing, are described. A model of recruitment combined with accentuated high-frequency loss is used to study speech perception in noise. It suggests an explanation for the previously reported ability of sensorineurals to tolerate the masking of speech by white noise at least as well as normals do, and for the special vulnerability of the speech perception of such deaf subjects to masking signals that have the spectral distribution of speech. A second model, that simulates the severely restricted dynamic range of hearing in profound deafness, predicts that this one factor can destroy the intelligibilty of amplified speech by making the perceived sound drop out at frequent intervals. Finally, a model that simulates reduced frequency-discriminating ability is described, as an example of a model that deals with aspects of perception other than loudness.

Envelope Coding in Auditory Nerve Fibers Following Noise-Induced Hearing Loss

“…It is possible that magnified masker fluctuations due to enhanced envelope coding could be a distraction that would reduce speech intelligibility. This idea is consistent with simulations of the effects of loudness recruitment with normal-hearing listeners, which showed that amplitude expansion reduced speech intelligibility, particularly in fluctuating background noises (Villchur 1977;Moore and Glasberg 1993;Moore et al 1995). Enhanced envelope coding is also likely to contribute to reduced ability of hearing-impaired listeners to detect temporal gaps between narrowband noises (Fitzgibbons and Wightman 1982;Glasberg et al 1987;Glasberg and Moore 1992).…”

Simulation of the effects of loudness recruitment and threshold elevation on the intelligibility of speech in quiet and in a background of speech

“…Provided the simulation is accurate, this makes it possible to study the effect of that aspect in isolation. This has been applied to several aspects of hearing impairment, including loudness recruitment (Villchur, 1974;Duchnowski, 1989;Glasberg and Moore, 1992) and reduced frequency selectivity (Villchur, 1977;Summers and A1-Dabbagh, 1982; Celmer and Bienvenue, 1987; Gagn6 and Erber, 1987;ter Keurs et al, 1992; Baer and Moore, 1993). This study is concerned with simulating the effect of loudness recruitment and the associated reduction in dynamic range on the ability to understand speech.…”

Relative contributions of spectral and temporal cues for phoneme recognition

“…Studies using spectral smearing have demonstrated detrimental effects of reduced spectral resolution on speech recognition (e.g., Villchur, 1977;ter Keurs et al, 1992ter Keurs et al, , 1993Moore, 1993, 1994;Boothroyd et al, 1996). However, the effect on speech recognition in quiet was hardly noticeable, even for smearing that simulated auditory filters six times broader than normal Moore, 1993, 1994).…”