Brunelle, M. (2009) Tone perception in Northern and Southern Vietnamese. Journal of Phonetics, 37, 79–96
Standard Vietnamese has a complicated tonal system that utilizes pitch height, contour, and voice quality. Given the fact that previous studies mostly focused on various acoustic measurements and dialectal differences have not been taken into account, this study aims to investigate Vietnamese tones perceptually, in the context of regional diversity.
The major tonal difference between Northern (standard) and Southern Vietnamese is that while the Northern dialect has six tones and utilizes glottalization as a cue, the Southern dialect has only five and does not use glottalization. The stimuli for the perception experiment, recorded by a Northern Vietnamese speaker, were later resynthesized so that different pitch contours (rising, falling, flat, and concave) were superimposed on syllables of three voice qualities – modal, slightly creaky, and glottalized. Listeners of the Northern and Southern dialects were asked to do tonal identification and goodness rating for all tokens. Perception results showed that when listening to the stimuli without synthesis, the two groups had similar identification rates. However, when synthesized stimuli were presented, there was a discrepancy between the Northern and Southern groups. In particular, listeners of the Northern dialect paid a lot of attention to voice quality, in which the presence of slight creakiness or glottalization greatly determined their identification. On the contrary, southerners did not view voice quality as crucial, and to them, glottalization sounded unnatural for most of the time. The analyses further revealed that for southerners, the pitch slope played a more important role for their identification than voice quality.
The cross-dialectal perception data presented in this study suggested that different strategies were adopted for perception by speakers of different dialects. Irrespective of dialectal difference, speakers could still make adjustments for accurate perception when listening to different dialects. The author thus concluded that a more sound tonal model should be proposed to account for both the variation and adaptability made between dialects.
No comments:
Post a Comment