Formant Characteristics of Vowels Produced by Mandarin Esophageal Speakers

Manwa L. Nga , Hanjun Liub

With the use of the pharyngoesophageal (PE) segment as the neoglottis in esophageal speakers, their voice quality appears to be significantly different from that of normal laryngeal speakers. Many researchers have investigated the perceptual, visual, temporal, acoustic, and aerodynamic aspects of voice quality associated with neoglottal vibration. However, relatively few studies examined the vocal tract transmission function in Mandarin esophageal speakers.
The present study attempted to determine if vocal tract resonance is changed after laryngectomy in esophageal speakers. The first three formant frequencies of vowels /?, ?, i, ?, u, y/ produced by seven superior esophageal and seven laryngeal speakers of Mandarin were obtained. All vowels were produced at the high level tone for at least one second. The voice signals were digitized at 20 kHz at 16bits/sample by using Praat. To obtain the formants, the waveform and wide-band spectrogram (bandwidth = 20 Hz) of the sample were displayed, and the medial 80% was carefully marked. Formant frequency values were measured by using the LPC algorithm (window size = 25 ms) of Praat. For each formant, the values were averaged, and data obtained from esophageal speakers were compared with that obtained from laryngeal speakers.
Preliminary data indicated that esophageal speakers were associated with significantly higher formant frequencies (F1, F2, and F3) than laryngeal speakers. This is consistent with findings on English, Dutch, and Spanish esophageal speakers reported previously. Yet, with the exception of F3, the formant differences between esophageal and laryngeal speakers appeared to be varying from vowel to vowel.

a Division of Speech and Hearing Sciences, University of Hong Kong, Hong Kong, SAR China
b Department of Communication Sciences and Disorders, Northwestern University, Evanston, IL, 60208, USA.
Manwa Lawrence Ng, Ph.D.
Associate Professor
Division of Speech and Hearing Sciences
Faculty of Education
University of Hong Kong
852-28590582
Manwa.L.Ng@hku.hk