Dissertations, Theses, and Capstone Projects

Date of Degree

6-2024

Document Type

Dissertation

Degree Name

Ph.D.

Program

Speech-Language-Hearing Sciences

Advisor

Douglas H. Whalen

Committee Members

Christine H. Shadle

Ronald C. Scherer

Mira Goral

Subject Categories

Communication Sciences and Disorders | Music Pedagogy | Music Performance | Music Practice | Speech and Hearing Science

Keywords

sopranos, singing voice, singing voice science, second passaggio transition, acoustics of singing voice, articulation of singing voice

Abstract

This dissertation describes three experiments designed to explore the acoustic and articulatory activity of a perceptible change in the soprano voice known as the secondo passaggio, or second transition, which typically occurs for the [ɑ] vowel between D♮5 (587 Hz) and F♮5 (698 Hz). Acoustically, the transition involves a shift in resonance strategy from second resonance tracking of the second harmonic (fR2/2fo) to first resonance tracking of the fundamental frequency (fR1/fo). To quantify the acoustics of the transition, we employed the measure L1-L2, the radiated sound level (in dB) difference between the first two harmonics. To measure articulation associated with the transition, we used ultrasound of the tongue with concomitant optical tracking.

In the first experiment, we tested the validity and reliability of L1-L2 to quantify the presence (deemed "techniqued singing") or absence (deemed "untechniqued singing") of the transition on chromatic scales between A♭4 and A♭5. Our participant was a classically trained soprano who produced consistent renditions of both types of singing. Results of techniqued singing revealed a change from negative to positive values in L1-L2 at a consistent pivot point between E♭5 and E♮5 while untechniqued singing exhibited only negative values. Analysis of vibrato cycle showed that L1-L2 values alternated between negative and positive values at the pivot point depending on the location of fo in the vibrato cycle; L1-L2 values transitioned to positive values at vibrato maxima a whole tone earlier than at vibrato minima.

The second experiment was a combined acoustic/articulatory investigation that added two additional participants (one who consistently made the transition and one who did not) in addition to the participant of the first experiment. Acoustic results of chromatic scales and glissandi between C♮5 and G♮5 (523 to 784 Hz) replicated the acoustic change from negative to positive values of L1-L2 in techniqued singing and revealed either the absence of a change or inconsistency in the measure in untechniqued singing. Articulatory results showed that techniqued singing exhibited increased lip aperture, lowered mandibular position, and tongue advancement with increased arching, whereas untechniqued singing showed marked invariance in articulatory measures as well as an overall high position of the tongue in the vocal tract.

The third and main experiment expanded the acoustic/articulatory study to 17 sopranos; 9 were judged by the principal investigator to be techniqued and 8 to be untechniqued. These judgments were confirmed by a perceptual experiment. Results showed that sopranos who achieved values of L1-L2 greater than 15 dB SPL at G♮5 exhibited three acoustic characteristics: a consistent change from negative to positive values of the L1-L2 measure early in the chromatic scale (between D♮5 and E♭5), minimal variability within musical note, and a steep rate of change in the acoustic measure over the course of the scale. Articulatory analysis of eight measures from 2-D sagittal contours of the tongue and hard palate as well as lip and jaw position showed that singers with high values of L1-L2 at G♮5 exhibited large anterior oral cavities, whereas those who never made the change exhibited small anterior cavities. A linear mixed effects statistical analysis revealed that all eight articulatory measures at C♮5 were significant predictors of L1-L2 values at G♮5. The articulatory factor that best distinguished the high-L1-L2 group from the low-L1-L2 group was the size of a triangular area between two points on the hard palate and the most anterior point of the tongue.

Despite long held beliefs that the second passaggio transition is accomplished principally via articulatory adjustments to the vocal tract to encourage fR1/fo tracking, our analysis revealed that articulatory changes alone could not account for fR1/fo tracking or the overall increases in L1-L2. Instead, we propose a dual mechanism hypothesis in which a combination of articulation and glottal source mechanics contribute to the L1-L2 change through the transition. Specifically, though modification of the vocal tract is important for achieving the L1-L2 change, it is not to increase L1 through fR1/fo tracking. Instead, the articulatory changes are employed to decrease fR2, discourage fR2/2fo tracking, and lower L2. To increase L1, we suggest that increased glottal airflow is required, not fR1/fo tracking. To be clear, though fR1/fo tracking is still the desired goal at the highest frequencies of the soprano voice, present acoustic theories hold that it would not occur until fo increases enough to meet fR1, which is likely above 800 Hz for [ɑ] in high female voices. However, our results show that the transition to a dominant fundamental begins well before fR1/fo tracking is possible.

In sum, our research concludes that a successful second passaggio transition on [ɑ] involves two distinct parts. Part 1 occurs in the vicinity of 600 Hz and requires two distinct elements: (i) increased glottal airflow, which leads to increases in L1, and (ii) concomitant articulatory maneuvers that increase the anterior vocal tract airspace, lower fR2, and subsequently decrease L2. Part 2 of the transition involves tracking of fR1 with fo once fo increases high enough to meet the first resonance frequency of [ɑ], which would be above 800 Hz in sopranos. Once fo increases high enough to match fR1, tracking of the first resonance with the fundamental creates a powerful standing wave that can be sustained for the highest frequencies of the soprano voice.

Download

Share

COinS