Effects of Hearing Aid Amplification on Robust Speech Coding

摘要 助听器能够恢复听力障碍者的一些听力，但背景噪音仍然是一个严重的问题。不幸的是，人们对语音在听觉系统中是如何编码的知之甚少，尤其是在装有假肢放大器的受损系统中。更好地了解助听器对噪声中语音编码的影响可能会为这些设备的设计带来新的见解。越来越多的证据表明，跨纤维神经编码对语音感知很重要，但还没有研究将时空编码与助听器放大联系起来。 这项拟议的研究的目的是在听神经水平上表征助听器如何影响噪声中的元音编码，并评估一种旨在增强跨纤维编码的新型助听器解决方案。将在安静和存在背景噪声的情况下比较速率-位置、时间-位置和时空编码方案的强度。指导这项工作的一般假设是，尽管当前的频率相关增益策略不能改进时空编码，但基于生理相位效应的新型放大技术可以改进时空编码。 两种助听器算法对噪声(相对于正常和受损)元音编码的影响将被量化：线性助听器算法和多通道宽动态范围压缩算法。此外，还将评估全通滤波器增强受损时空图案的能力。提出了两种相位校正算法：一种是以元音共振峰为中心的全通滤波，另一种是以各次谐波分量为中心的全通滤波。 建议的神经生理学实验将提供关于助听器放大对噪声中元音的听神经编码的影响的有价值的数据。这些数据将有助于理解放大的效果，并设计助听器，使患者能够更好地理解噪音中的语音。

Abstract Hearing aids are able to restore some hearing abilities for people with auditory impairments, but background noise remains a significant problem. Unfortunately, very little is understood about how speech is encoded in the auditory system, particularly in impaired systems with prosthetic amplifiers. A better understanding of the effects of hearing aids on speech coding in noise could lead to new insights for the design of these devices. There is growing evidence that across-fiber neural coding is important for speech perception, but there is no research that relates spatiotemporal coding and hearing aid amplification. The goal of the proposed research is to characterize how hearing aids affect vowel coding in noise at the level of the auditory nerve, and to evaluate a novel hearing aid solution that is designed to enhance across- fiber coding. The strength of rate-place, temporal-place, and spatiotemporal coding schemes will be compared in quiet and in the presence of background noise. The general hypothesis guiding the proposed work is that, although current frequency-dependent gain strategies do not improve spatiotemporal coding, novel amplification techniques based on physiological phase effects can improve spatiotemporal coding. The effects of two hearing aid algorithms on vowel coding in noise (relative to normal and impaired) will be quantified: a linear hearing aid algorithm and a multichannel wide-dynamic-range compression algorithm. In addition, the ability of all-pass filters to enhance impaired spatiotemporal patterns will be evaluated. Two phase-correction algorithms are proposed: one with all-pass filters centered on the formants of a vowel, and another with all-pass filters centered on each harmonic component. The proposed neurophysiological experiments will provide valuable data about the effects of hearing aid amplification on auditory nerve coding of vowels in noise. These data will be useful for understanding the effects of amplification and designing hearing aids that allow patients to better understand speech in noise.