ABSTRACT
This part introduction presents an overview of the key concepts discussed in the subsequent chapters. The part reviews experiments in which a particular perceptual consequence of sensorineural hearing loss was simulated in normal-hearing listeners, to examine its effect on speech recognition in isolation. It demonstrates that the accuracy of the Articulation Index in predicting speech recognition scores of hearing-impaired subjects is improved substantially by including several modifications related specifically to effects of sensorineural hearing loss, as proposed by J. Fletcher. The part suggests that residual differences between SII predictions and observed speech recognition scores are more strongly related to magnitude of hearing loss and speech level than to listeners’ frequency and temporal resolution. It shows that conclusions regarding the accuracy of audibility-based speech recognition predictions are highly dependent on the specific methods used to compute the index and develop the transfer function.
