Perception of Voicing Accuracy in Children's Speech

Thursday, 11:00am to 1:00pm
Seabright
Poster 1
Voice onset time (VOT) is the interval between the release of vocal tract occlusion and the onset of glottal vibration. VOT has frequently been used as the method for investigating voicing development in syllable-initial stop consonants (Lisker & Abramson, 1964); however, the majority of the research regarding the perceptual identification and discrimination of speech sounds related to VOT has used synthetic speech stimuli for perceptual tasks. The stimuli have been systematically altered to judge the impact of phoneme boundaries, lag time, lead time, and overall accuracy of perception. The purpose of the present study was to explore the implications of natural speech on discrimination of voiced and voiceless cognate pairs produced by 2-3 year-old monolingual, English speaking children. A subset of previously recorded single word targets from Hitchcock & Koenig, (2013) were re-measured to fit a continuum of VOT values ranging from 0 -100.99 ms. VOT categories were created in 10 millisecond intervals between 0.00ms and 100.99ms (0.00-10.99ms, 11.00-20.99, etc.). Two exemplars (voiced/voiceless) were randomly chosen from the 10 children in original study. Each exemplar corresponded to preset VOT intervals (200 word sample) randomly presented 2x for the listener. Twenty monolingual, English speaking speech pathology graduate students were asked to distinguish between voiced/voiceless bilabial and alveolar cognates in 400 CV words (boo/pooh, doe/toe). Accuracy and response times were cataloged. Preliminary findings indicate great than 90% accuracy between cognate pairs suggesting that adult listeners can easily discriminate the voicing contrast in the speech of children as young as 2-3 years old.
Track: 
Other