Importance of low-frequency on understanding interrupted speech: Sentence recognition of hearing-impaired (HI) and normal-hearing (NH) listeners

Abstract

In a recent study of HI listeners’ speech perception in noise (Jin and Nelson, 2004), two main factors, low-frequency audibility and auditory filter bandwidths, were highly related to amplified sentence recognition in modulated noise. Nine young adult listeners with sensorineural hearing loss and eight young NH listeners as controls participated in the series of experiments. Amplified speech recognition of the HI listeners was equal to that of the NH listeners in quiet and in steady noise, but was significantly poorer in modulated noise. Thus, even with adequate amount of amplification, HI listeners experienced significantly less masking release from the modulated maskers. The results indicated that those listeners with greatest hearing losses in the low frequencies were poorest at speech recognition in modulated noise. Additional results from HI and NH listeners will be presented, in which we systematically vary the audibility of different frequency regions of speech by filtering. Sentences are interrupted by either speech-shaped noise or silence gap while measuring the percent of sentence recognition. The purpose of the current study is to examine the importance of low-frequency audibility for auditory segregation/integration of interrupted speech.