Audio quality perception of hearing-impaired listeners in complex acoustic environments

The effect of complex acoustic environments (CAEs), typically comprising target and interfering sound sources as well as room reflections, on speech reception of hearing impaired (HI) listeners have been examined in several studies. However, only little is known about audio quality perception of HI listeners in CAEs. Thus, this study assessed detection thresholds and supra-threshold audio quality ratings of listeners with very mild and moderate hearing loss (HL) for several distortions applied to speech and pink noise: non-linear saturation, spectral ripples, level differences, and spatial position offsets. The stimuli were presented in acoustical scenes that differ in their complexity by manipulating room size in conjunction with reverberation time, and the number and spatial position of interfering sound sources. The strongest differences between listeners with very mild and moderate HL were observed in the presence of interfering sounds. In such situations, listeners with moderate HL had consistently higher distortion detection thresholds than listeners with very mild HL. Moreover, they rated audio quality lower for the masked than for the unmasked distorted targets, indicating difficulties in separating the target from the maskers. Significant correlations were found between the listeners’ pure tone average (PTA) and distortion detection thresholds in situations with maskers. Thus, PTAs seem to be a suitable predictor for distortion thresholds of HI listeners in CAEs. The effect of reverberation strongly depended on the target (speech or pink noise) and the type of distortions.