Sentence intelligibility during segmental interruption and masking by speech-modulated noise

Abstract

Amplitude modulation from a competing talker can interfere with a listener's ability to process informative speech cues. The current study investigated how noise modulated by the wideband temporal envelope of a single competing talker impacts the contribution of consonants and vowels to the intelligibility of the target sentence. High variability speech materials were used, including talkers from different dialect regions. Young normal-hearing, older normal-hearing, and older hearing-impaired listeners completed speech recognition tests. All listeners received spectrally shaped speech matched to their individual audiometric thresholds to ensure sufficient audibility. Preliminary results demonstrated similar performance among the listener groups. When the modulated masker was interrupted, performance in consonant and vowel conditions was similar. However, sentence intelligibility in continuous single-talker-modulated noise was higher when vowels in the target sentence were preserved, as compared to consonants. Thus, masker continuity may facilitate source segregation for vowels more than for consonants. A second experiment that varied masker modulation rate for younger adults with normal hearing showed that modulation rate had more effect on processing cues conveyed by consonants than by vowels. Implications for hearing-impaired and older adults will be discussed. [Work supported by grants from NIH/NIDCD and ASHA.]