Native and nonnative speakers' perception of syllable initial voiced stop consonants in three contexts: Non-manipulated, edited to remove VOT and edited to remove formant transitions.

摘要

The purpose of this study was to evaluate how accurately initial voiced plosive consonants /b, d, g/ produced before the cardinal vowel /a/ were identified by native (n=16) and nonnative (n=14) speakers of English. Plosives were presented in non-manipulated recordings, recordings in which the formant transitions (FT) were removed, and recordings in which the voice onset time (VOT) was removed. A two-group repeated measures descriptive study was implemented. The results revealed that both native and nonnative speakers of English identified non-manipulated recordings of the syllables with 99% accuracy; the removal of VOT and FT individually both negatively impacted the listeners' ability to identify the plosive; neither piece of acoustic information appeared to drastically improve or hinder identification more than the other. Both groups had more trouble identifying the velar phoneme /g/ in edited stimuli, with nonnative speakers having significantly more trouble identifying /g/ in edited stimuli than native speakers.