The effects of phonological content, sentence context, and vocoding on voice cue perception

Abstract

Recently, we showed the effect of lexical content on just-noticeable-differences (JNDs) in voice pitch (F0) and vocal-tract length (VTL) voice cue perception. Participants performed an adaptive 3AFC task with non-vocoded and vocoded auditory stimuli. When presented with words, participants showed smaller VTL JNDs compared to time-reversed words, also when vocoding was applied. These outcomes inspired two follow-up studies using similar methods. The first study compared words, time-reversed words, and non-words, to investigate if the benefit is related more specifically to lexical (words) and/or phonological (non-words) content. Irrespective of vocoding, reversed words had a detrimental effect on VTL perception, replicating earlier findings. However, VTL JNDs with non-words did not significantly differ from those with words, suggesting linguistic content benefits for VTL perception at a phonological level. The second study compared JNDs with words and sentences and showed, again regardless of vocoding, a benefit in processing a full sentence compared to a single word in both F0 and VTL perception, suggesting that the amount of acoustic speech information and/or semantic context led to increased voice cue sensitivity. These results improve our understanding of the interactive speech and voice perception processes, inspiring rehabilitation tools for CI users with limited access to voice information.