Croghan Naomi B H, Arehart Kathryn H, Kates James M
Department of Speech, Language, and Hearing Sciences, University of Colorado, Boulder, Colorado, USA.
Ear Hear. 2014 Sep-Oct;35(5):e170-84. doi: 10.1097/AUD.0000000000000056.
Current knowledge of how to design and fit hearing aids to optimize music listening is limited. Many hearing-aid users listen to recorded music, which often undergoes compression limiting (CL) in the music industry. Therefore, hearing-aid users may experience twofold effects of compression when listening to recorded music: music-industry CL and hearing-aid wide dynamic-range compression (WDRC). The goal of this study was to examine the roles of input-signal properties, hearing-aid processing, and individual variability in the perception of recorded music, with a focus on the effects of dynamic-range compression.
A group of 18 experienced hearing-aid users made paired-comparison preference judgments for classical and rock music samples using simulated hearing aids. Music samples were either unprocessed before hearing-aid input or had different levels of music-industry CL. Hearing-aid conditions included linear gain and individually fitted WDRC. Combinations of four WDRC parameters were included: fast release time (50 msec), slow release time (1,000 msec), three channels, and 18 channels. Listeners also completed several psychophysical tasks.
Acoustic analyses showed that CL and WDRC reduced temporal envelope contrasts, changed amplitude distributions across the acoustic spectrum, and smoothed the peaks of the modulation spectrum. Listener judgments revealed that fast WDRC was least preferred for both genres of music. For classical music, linear processing and slow WDRC were equally preferred, and the main effect of number of channels was not significant. For rock music, linear processing was preferred over slow WDRC, and three channels were preferred to 18 channels. Heavy CL was least preferred for classical music, but the amount of CL did not change the patterns of WDRC preferences for either genre. Auditory filter bandwidth as estimated from psychophysical tuning curves was associated with variability in listeners' preferences for classical music.
Fast, multichannel WDRC often leads to poor music quality, whereas linear processing or slow WDRC are generally preferred. Furthermore, the effect of WDRC is more important for music preferences than music-industry CL applied to signals before the hearing-aid input stage. Variability in hearing-aid users' perceptions of music quality may be partially explained by frequency resolution abilities.
目前关于如何设计和适配助听器以优化音乐聆听的知识有限。许多助听器使用者会听录制音乐，而录制音乐在音乐行业中通常会经过压缩限制（CL）处理。因此，助听器使用者在听录制音乐时可能会体验到双重压缩效果：音乐行业的CL和助听器的宽动态范围压缩（WDRC）。本研究的目的是研究输入信号特性、助听器处理以及个体差异在录制音乐感知中的作用，重点关注动态范围压缩的影响。
一组18名有经验的助听器使用者使用模拟助听器对古典音乐和摇滚音乐样本进行配对比较偏好判断。音乐样本在输入助听器之前要么未经过处理，要么具有不同程度的音乐行业CL。助听器条件包括线性增益和个性化适配的WDRC。包括了四个WDRC参数的组合：快速释放时间（50毫秒）、慢速释放时间（1000毫秒）、三通道和18通道。聆听者还完成了几项心理物理学任务。
声学分析表明，CL和WDRC降低了时间包络对比度，改变了整个声谱的幅度分布，并平滑了调制谱的峰值。聆听者的判断显示，快速WDRC在两种音乐类型中都是最不受欢迎的。对于古典音乐，线性处理和慢速WDRC同样受欢迎，通道数量的主要影响不显著。对于摇滚音乐，线性处理比慢速WDRC更受欢迎，三通道比18通道更受欢迎。重度CL在古典音乐中最不受欢迎，但CL的量并没有改变两种音乐类型中WDRC偏好的模式。根据心理物理学调谐曲线估计的听觉滤波器带宽与聆听者对古典音乐偏好的变异性相关。
快速、多通道的WDRC通常会导致音乐质量较差，而线性处理或慢速WDRC通常更受青睐。此外，对于音乐偏好而言，WDRC的影响比在助听器输入阶段之前应用于信号的音乐行业CL更为重要。助听器使用者对音乐质量感知的变异性可能部分由频率分辨能力来解释。
