Effect of Vowel Type on Aerodynamic Measures of Estimated Subglottal Pressure and Airflow

The purpose of this study was to investigate the potential effect of factors such as tongue height and advancement and vocal tract constrictions on aerodynamic measures of voice (mean airflow [MAF] and estimated subglottal pressure [Psub]). A key question was whether the choice of elicitation vowel would have a substantial impact on normative expectations and subsequent clinical decision making that may be based on measures of MAF or Psub.

Forty-three nondysphonic male (n = 13) and female (n = 30) participants between the ages of 18 and 30 years were tested using the Phonatory Aerodynamic System (Model 6600, Pentax Medical Inc.). MAF was measured from sustained vowels /i/, /u/, and /α/ in their comfortable pitch and loudness using the comfortable sustained phonation protocol. MAF and Psub were also measured from repetitions of the syllable /pV/, where V was the target vowel (/i, u, α/. Measurement of vocal fundamental frequency (F0) during vowel production in the aforementioned protocols was also measured. The order of sustained vowel versus syllable production tasks and the target vowels were counterbalanced across subjects.

Analysis of MAF in comfortable sustained phonation showed a significant main effect of vowel type (small effect size), with post hoc analysis indicating that the MAF observed for /i/ was significantly lower than that observed for /α/ (0.139 vs. 0.152 L/s, respectively). MAF for the vowel /u/ fell in between (0.147 L/s), with differences in MAF between /i/ versus /u/ and /α/ versus /u/ being nonsignificant. In contrast, the analysis of MAF in syllable context showed no significant main effect of vowel. The main effect of vowel was significant for measures of Psub obtained in the syllable context (small effect size), with syllables including the high-impedance vowel /i/ observed to have significantly higher Psub compared with syllables including the low-impedance vowel /α/ (7.96 vs. 7.54 cm H₂O, respectively). Mean Psub for the syllables including /u/ was again observed to fall in between (7.64 cm H₂O), with differences in Psub between /i/ versus /u/ and /α/ versus /u/ being nonsignificant. Measures of MAF and Psub from different vowel contexts were highly correlated. Vocal F0 was not observed to be a significant predictor of MAF or Psub.

Although the generally small effect sizes for the effect of vowel type on aerodynamic measures indicate that this finding may be of little practical significance, the results of this study support the view that vowel type has a predictable and measurable effect on measures of MAF (in comfortable sustained phonation) and Psub in syllable context. Supraglottal impedance changes secondary to articulatory differences between vowels are primarily responsible for the differences between vowels on MAF and Psub versus F0-related changes in laryngeal activity that may accompany different vowel productions. The results of this study indicate that vowel type should not have any substantial effect on normative expectations for MAF or Psub, and linear regression equations may be used to convert aerodynamic measures of MAF and Psub between different vowels with relatively little error.