The Effect of Cue-enhancement on Consonant Intelligibility in Noise: Speaker and Listener Effects

Our previous work (Hazan & Simpson, 1998) has shown that increasing the salience of perceptually important regions of nonsense word and sentence materials aids speech perception in noise. This study aimed to extend these findings by investigating the robustness of these enhancement techniques in improving consonant intelligibility for a range of different speakers and for groups of listeners with different language backgrounds. In Experiment 1, nonsense vowel-consonant-vowel (VCV) tokens produced by two female and two male speakers without phonetic training were annotated to highlight segments of the signal corresponding to the consonant constriction/occlusion and vowel onset/offset regions. These regions were selectively amplified to enhance the cues they contained, then combined with speech-shaped noise at 0 dB SNR and presented to normally hearing listeners. Significant improvements in intelligibility were found for all speakers although the extent of the improvement varied across speakers. In Experiment 2, a subset of these stimuli were presented to two groups of learners of English--a Japanese-L1 group and a Spanish-L1 group--and a new group of native-English controls. Results showed a significant effect of enhancement in all listener-groups and similar speaker effects for the non-native and native English listeners. Error patterns were related to the distance between the phonological systems of listeners' L1 and L2.