Auditory-model based assessment of the effects of hearing loss and hearing-aid compression on spectral and temporal resolution

Most state-of-the-art hearing aids apply multi-channel dynamic-range compression (DRC). Such designs have the potential to emulate, at least to some degree, the processing that takes place in the healthy auditory system. One way to assess hearing-aid performance is to measure speech intelligibility.

However, due to the complexity of speech and its robustness to spectral and temporal alterations, the effects of DRC on speech perception have been mixed and controversial. The goal of the present study was to obtain a clearer understanding of the interplay between hearing loss and DRC by means of auditory modeling.

Inspired by the work of Edwards (2002), we studied the effects of DRC on a set of relatively basic outcome measures, such as forward masking functions (Glasberg and Moore, 1987) and spectral masking patterns (Moore et al., 1998), obtained at several masker levels and frequencies. Outcomes were simulated using the auditory processing model of Jepsen et al. (2008) with the front end modified to include effects of hearing impairment and DRC.

The results were compared to experimental data from normal-hearing and hearing-impaired listeners.