Acoustic Analysis of Nasalization in Mandarin Prenasal Vowels Produced by Wenzhou and Rugao Speakers

This study investigates the effects of first dialect, age, vowel type, nasal type and vowel timing on the nasalization of prenasal vowels in second dialect standard Mandarin produced by respectively Wenzhou and Rugao speakers. The nasalization was quantified by the values of the acoustic correlates A 1-P 1 (dB) and A 1-P 0 (dB) in the amplitudes of spectral frequencies of the vowels. The statistical results of mixed-effects linear regression model reveal different patterns of nasalization of prenasal vowels in Mandarin produced by young and middle-aged speakers of Wenzhou and Rugao dialects. The degree of nasalization was found varying across vowel type and time course of the vowel but not nasal type. The appropriate acoustic parameter for testing nasalization of the mid vowel /ə/ remains uncertain because A 1-P 1 is supposed to be used for measuring nasalization of nonlow vowels while A 1-P 0 for nonhigh vowels [11, 12].