Acoustic and perceptual effects of WDRC amplification

描述(申请人提供)：本研究计划的总体重点是描述多通道宽动态范围压缩(WDRC)放大对语音清晰度的影响。助听器(以及后来的听力康复计划)现在都基于这项技术，预计它将比传统的放大策略提高语音清晰度。目前尚不清楚以前的研究中看到的多通道WDRC的不一致益处是由于助听器未能保存和传输声学信息，还是由于听者无法处理传输的信息。我们假设，信号可听度和/或时间线索的变化可以通过测量放大系统输出的语音特征来描述，并将部分解释不同受试者的清晰度差异。很少有人尝试通过实际测量传输给听者的语音的声学特性(即，在助听器的输出处)来理解助听器的效果。使用在放大系统的输出处测量的记录的自然语音令牌，我们建议识别被多通道WDRC增强或干扰的特定声学提示(例如，上升时间、频谱峰值)。此外，我们打算测量这些提示因压缩参数(释放时间、通道数量)的变化而改变的程度。我们将把声音信号的变化与清晰度联系起来。最后，我们将确定WDRC对语音清晰度的影响是否作为年龄的函数而不同，以及与年龄相关的性能差异(如果存在)是否与信号的时间特征的变化有关。

DESCRIPTION (provided by applicant): The overall focus of this research program is to describe the effects of multichannel wide-dynamic range compression (WDRC) amplification on speech intelligibility. Hearing aids (and by extension, hearing rehabilitation programs) are now based on this technology, which is expected to improve speech intelligibility over conventional amplification strategies. At present it is unclear whether the inconsistent benefit from multichannel WDRC seen in previous studies was due to a failure of the hearing aid to preserve and transmit acoustic information, or to the listener's inability to process transmitted information. We hypothesize that changes in signal audibility and/or temporal cues can be described by measuring speech characteristics at the output of the amplification system and will partially account for differences in intelligibility across subjects. Few attempts have been made to understand the effects of hearing aids by actually measuring the acoustic characteristics of speech as transmitted to the listener (i.e., at the output of the hearing aid). Using recorded natural speech tokens measured at the output of the amplification system, we propose to identify specific acoustic cues (e.g., rise time, spectral peak) that are enhanced or disrupted by multichannel WDRC. Additionally, we intend to measure the extent to which these cues are altered by variations in compression parameters (release time, number of channels). We will relate changes in the acoustic signal to intelligibility. Finally, we will determine if the consequences of WDRC for speech intelligibility differ as a function of age and whether age-related performance differences, if they exist, are linked to the alteration of the temporal characteristics of the signal.