The effect of dialect variation and hearing aid compression type on speech recognition in hearing impaired listeners
Loading...
Date
2011-06
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Ohio State University. Department of Speech and Hearing Science
Abstract
It is well established that normal hearing listeners presented with speech spoken in an unfamiliar dialect demonstrate a reduction in understanding abilities. Although listeners with hearing-impairment are known to experience more struggles with understanding speech due to variability of the signal, little investigation of dialect understanding in hearing impaired listeners has occurred. Additionally, it is not known how amplification may assist the hearing impaired listener with understanding speech spoken in an unfamiliar dialect. Listeners with hearing impairment are likely to encounter speakers from different dialect areas on a regular basis in their daily lives. Therefore, it is important to establish how well hearing impaired listeners understand speech spoken in an unfamiliar dialect, and how hearing aids can be set to potentially maximize performance.
The present study examined the speech understanding abilities of hearing impaired listeners when presented with sentences spoken by a speaker from the Midlands dialect area, and the same sentences spoken by a speaker from the South dialect area. The sentences were presented in quiet and in noise. The performance of the hearing impaired listeners was compared with the performance of normal hearing counterparts. Results show that both groups of listeners performed poorer with the speech spoken by the non-native dialect speaker (i.e., the South dialect). The addition of background noise did not affect the normal hearing subjects’ performance. However, the hearing impaired listeners performed poorer on speech presented in noise, regardless of the dialect. The effect of hearing impairment and background noise interacted to create disproportionately poorer performance in the hearing impaired group in noise.
The hearing impaired listeners were fitted with hearing aids in order for speech understanding abilities to be examined with fast-acting and slow-acting compression. The hearing impaired listeners were presented with the sentences in quiet and noise while wearing binaural hearing aids fit with fast-acting and slow-acting compression. Results indicate that no one compression strategy appears to improve speech understanding abilities more than the other across all listeners tested. It is possible that a slight advantage is seen with the use of fast-acting compression, but this difference is not significant. Variability in scores across subjects and small sample size may have contributed to the lack of difference in performance due to compression setup. Subjectively, however, each subject remained consistent on the compression strategy which they preferred. This suggests the importance of taking listener preference into account when adjusting compression settings, even if only for subjective acceptance.