See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/317825890

Effects of Straw Phonation Through Tubes of Varied Lengths on Sustained

Vowels in Normal-Voiced Participants

Article · June 2017

DOI: 10.1016/j.jvoice.2017.05.015

CITATIONS READS

40 2,229

5 authors, including:

Jack J Jiang
University of Wisconsin–Madison
337 PUBLICATIONS 8,163 CITATIONS

SEE PROFILE

All content following this page was uploaded by Jack J Jiang on 22 September 2017.

The user has requested enhancement of the downloaded file.

 ARTICLE IN PRESS
Effects of Straw Phonation Through Tubes of Varied
Lengths on Sustained Vowels in Normal-Voiced
Participants
Randal D. Mills, Shawna Rivedal, Colten DeMorett, Grace Maples, and Jack J. Jiang, Madison, Wisconsin

Summary: Objective. To examine the immediate effects of straw phonation exercises in normal subjects while al-
tering the effective length of the vocal tract.
Study Design. A nonrandomized comparison of semi-occluded vocal tract length during straw phonation exercises
was carried out.
Methods. Oral pressure, mean airflow, aerodynamic resistance, and contact quotient were measured in 20 healthy
subjects immediately before and after straw phonation exercises. A short- and long-duration phonatory task was used
to examine the voice parameters during semi-occluded vocal tract exercises. These tasks involved repeating a vocal-
ization of the vowel /a/ through a tube. Each subject underwent the protocol using tubes of three different lengths (7.5 cm,
15 cm, and 30 cm) to allow for the effect of moving the outlet of the vocal tract distal to the mouth to be monitored.
Results. Oral pressure and aerodynamic resistance decreased significantly, contact quotient showed a decreasing trend,
and airflow increased significantly in pre- and postmeasurements 15 minutes after a long-duration phonatory task. Short-
duration tasks were found to have no effect on voice parameters.
Conclusion. The results present and validate a method to isolate the effect that the length of a semi-occluded vocal
tract has during straw phonation exercises.
Key Words: Straw phonation–Voice therapy–Vocal tract–Impedance–Human subject.

INTRODUCTION
Voice disorders impair work performance and quality of life, and
approximately 28 million workers in the United States require
their voice for employment.1 The prevalence of voice disorders
demands accessible and noninvasive vocal therapy.
Semi-occluded vocal tract (SOVT) therapy is a noninvasive
treatment for voice disorders that improves vocal economy, ef-
ficiency, and intensity. Variations include hand over mouth, lip
and tongue trills, Y-buzz, bilabial fricatives (eg, [β:]), nasal con-
sonants (eg, /m/, /n/, /ε/), resonance tubes, and straws.2 Mouth
semi-occlusions, such as bilabial fricatives, lip trills, /u/ vowels,
and straws produce resonant voice. Resonant voice is a target
for vocal production because it does not cause excessive me-
chanical stress to laryngeal tissues or result in poor harmonic
content. Further, it also provides a more effective conversion of
aerodynamic to acoustic energy.3
During phonation, the glottis is opened when the vocal tract’s
air column is accelerated by increased glottal flow, resulting in
a positive supraglottal and subglottal pressure. This coupling
results in a positive intraglottal pressure that pushes the vocal
folds apart.2–5 Supraglottal pressure becomes negative during the
closing phase, creating a suction that pulls the vocal folds apart.2–5
The strength of this push-pull interaction is determined by the
acoustic impedance of the vocal tract.2–5 Impedance consists of
resistance, which removes energy from a system, and reac-
tance, which stores and feeds energy into a system. Reactance
Accepted for publication May 19, 2017.
From the Division of Otolaryngology—Head and Neck Surgery, Department of Surgery,
University of Wisconsin-Madison School of Medicine and Public Health, Madison,
Wisconsin.
Address correspondence and reprint requests to Jack Jiang, 1300 University Avenue, 2745
Medical Sciences Center, Madison, WI 53706. E-mail: jiang@surgery.wisc.edu
Journal of Voice, Vol. ■■, No. ■■, pp. ■■-■■
0892-1997
Published by Elsevier Inc. on behalf of The Voice Foundation.
http://dx.doi.org/10.1016/j.jvoice.2017.05.015
can further be broken down into positive and negative compo-
nents. The positive component, inertive reactance, produces a
time-delayed buildup of supraglottal pressure, strengthening the
push-pull relationship that facilitates self-sustained oscillation.2–5
High inertance is therefore desired as it leads to greater ampli-
tude of vocal fold vibration which increases source strength in
the nonlinear source-filter interaction and allows for easier
phonation.6 A schematic which breaks down the components of
acoustic impedance can be seen in Figure 1.
Impedance matching seeks to lessen tissue collision and
thus cause improved voice production. The most efficient
transfer of power from the source-filter interaction, glottis to
lips, occurs when source impedance is matched to vocal tract
input impedance.7 Impedance is altered in straw phonation by
having a patient phonate through a tube that alters both the
effective length and outlet diameter of the vocal tract. These
alterations to the effective vocal tract can improve vocal
economy and vocal efficiency. Vocal economy is a measure-
ment involving voice output and intraglottal impact stress and
helps to maintain resonant voice and inertive reactance.8 High
vocal economy is obtained when voice output is high while
intraglottal impact stress remains low.8 Intraglottal impact
stress, which is defined as the impact force of the vocal folds
during phonation divided by contact surface area of the vocal
folds, is a common cause of vocal lesions.8–10 Vocal efficiency
is a quantitative measurement that involves the conversion of
subglottal energy to acoustic energy.11,12 With higher efficien-
cy, less effort is needed to produce vocalizations.11,12 Improving
both vocal economy and vocal efficiency can therefore lead to
healthier voice production.
The effects of various dimensions of tubes used for straw pho-
nation have been studied in both excised laryngeal and other
simulated systems. Smith and Titze developed an experimental
setup to characterize the effects of tubes of various diameters
 ARTICLE IN PRESS
2 Journal of Voice, Vol. ■■, No. ■■, 2017
FIGURE 1. Schematic of the components of acoustic impedance. The inertive and compliant components of reactance depend on frequency of
the acoustic signal and the proportionality constants: inertance and compliance, respectively. I, inertance; C, compliance; ρ, density of air column;
L, length; A, cross-sectional area; c, speed of sound in air.
and lengths.13 Using four lengths divided equally from 3 cm to
24 cm, they were able to characterize differences in airflow (AF),
pressure, and resistance for these different tubes. Conroy et al.
used lengths within a similar range, specifically 7.5, 15, and
30 cm, in an excised laryngeal setup.14 They found a signifi-
cant decrease in phonation threshold pressure from control, when
using the 30-cm tube. These studies have been referenced to de-
termine the range of interest when evaluating the effect of the
length of the tube used for straw phonation.
Studies have examined the effects of different SOVT dimen-
sions across multiple therapy methods in human subjects.15,16
Further, no human subject studies have altered only length or
diameter while leaving the other constant. By evaluating only
the length of the SOVT used for straw phonation, a more precise
quantification of the physical changes observed may be found.
The purpose of this study is therefore to observe the isolated
effects of tube elongation on straw phonation exercises in healthy-
voiced subjects. We hypothesized that vocalizations following
SOVT exercises will be characterized by differences in aero-
dynamic and electroglottograph (EGG) parameters. Extended
duration SOVT tasks will result in significantly greater changes
in parameters than short-duration tasks.
METHODS
Subjects
A total number of 20 subjects (10 men, 10 women) volun-
teered for this study. Participants included undergraduate and
graduate students from the University of Wisconsin-Madison.
Recruitment was carried out by placing fliers in campus facili-
ties. The inclusion criteria were as follows: (1) nonsmoking, (2)
above the age of 18, and (3) having no history of diagnosed voice
disorders or other pertinent medical conditions (polyps, nodules,
laryngitis, etc). Before data collection, the Institutional Review
Board of the University of Wisconsin-Madison approved all
procedures.
Data collection
Data collection of this study consisted of a short- and long-
duration phonatory task, both of which are minimal risk tasks.
There were three different tube lengths evaluated in this study:
7.5 cm, 15 cm, and 30 cm PolyVinyl Chloride tubes, each with
a diameter of 2 cm. For each tube length, both the short- and
long-duration tasks were completed on the same day. However,
the three lengths were completed on three separate days within
a period of 2 weeks. This isolated the effects of each condi-
tion, while also ensuring each subject’s quality of voice was
similar across their 3 days of data collection.
Before initially starting the short-duration task, each patient
was given a training session on the equipment and the vocal-
izations they would be performing. This was carried out by
performing a demo of data collection for each subject. After this,
subjects were asked to get comfortable with the equipment by
performing test vocalizations. Once they were consistently able
to perform vocalizations with similar pitch and loudness, they
were allowed to move on to the short-duration task.
The short-duration task required the participant to perform 12
repetitions of the vowel sound /u/ through the SOVT for 5 seconds
in duration, with a 3-second break between each trial. The long-
duration version of this task required the subject to repeat the
short-duration task 10 times sequentially with a 30-second break
between iterations. Throughout the length of the study, partici-
pants were instructed to produce the pitch and loudness that is
most comfortable while directing all the sound through the SOVT.
Measurements of oral pressure, mean AF, mean aerody-
namic resistance (AR), and contact quotient (CQ) were taken
before and after each task for each tube length. EGG measure-
ments were obtained using a two-channel electroglottograph with
a microphone preamplifier (EG2-PCX2, Glottal Enterprises, Syr-
acuse, NY), whereas aerodynamic measurements were obtained
with the Phonatory Aerodynamic System (PAS Model 6600,
KayPENTAX, Montvale, NJ). No custom equipment or soft-
ware was used to collect data. Subjects were asked to complete
 ARTICLE IN PRESS
Randal D Mills, et al Effects of Straw Phonation Through Tubes of Varied Lengths
FIGURE 2. Flow chart of protocol used for data collection. Note: Short
and long refer to the duration of the phonatory task.
three trials of seven vocalizations into the PAS system with the
EGG electrodes secured around their larynx using a Velcro collar
(Glottal Enterprises, Syracuse, NY). Each vocalization was a
2-second /pa/ followed by 2 seconds of rest. Subjects were also
asked to indicate their level of effort and discomfort after each
task on a written scale of 0–10, with 0 being no amount of effort/
discomfort and 10 being extreme effort/discomfort. Also, there
was a 15-minute rest period between postmeasurements of the
short-duration task and premeasurements of the long-duration
task to isolate the effects of task duration. The procedure used
for data collection is shown in Figure 2.
DATA ANALYSIS
Values were averaged both within each measurement trial and
period. The data obtained through the seven vocalizations within
each trial were averaged. Further, the data obtained from the three
trials were averaged as well to obtain one value for each pa-
3
rameter per measurement period. Aerodynamic parameters were
compared using a two-way repeated measures analysis of vari-
ance in SigmaPlot 11.0 (Systat Software, Inc., Chicago, IL). This
statistical model accounted for the random effects of subjects
and the fixed effects of multiple measurement conditions and
tube elongations. It specifically tested the interaction effect
between tube length and measurement conditions, as well as the
main effects of tube length and measurement conditions on oral
pressure, AF, AR, and CQ. The measurement conditions are
defined as the pre- and postmeasurements taken for the short-
and long-duration tasks, as well as those taken across the entire
duration of each individual data collection period. Therefore, the
degree of freedom of measurement conditions is 4. The tube
lengths are 7.5, 15, and 30 cm, having a degree of freedom of
2. Post hoc Bonferroni tests were run on parameters that showed
significant interaction or main effects (P < 0.05) to obtain pairwise
comparisons. The Levene test was run to evaluate equal vari-
ance. Graphs were plotted for all parameters across task condition
and length to visualize trends.
RESULTS
Oral pressure (cmH2O)
A trend toward an interaction effect between measurement con-
ditions and tube length was found (P = 0.097). Further, there was
a significant main effect of measurement conditions (P = 0.020)
and a nonsignificant main effect of tube length (0.703). Evalu-
ating the main effect of measurement conditions, a significant
decrease was found across pre- and posttreatment measure-
ments (P = 0.020). No significant changes were observed across
the measurements taken for short- and long-duration tasks with
respective P values of 0.615 and 1.000. The premeasurements
of short- and long-duration tasks were compared and a nonsig-
nificant difference of means of 0.098 cmH 2 O was found
(P = 1.00). The oral pressure data passed the equal variance test
(P = 0.867). Power values for data compiled according to the
interaction effect, measurement conditions versus tube length
(0.323), and the main effects, measurement conditions (0.593)
and tube length (0.05), were evaluated with an alpha of 0.05.
Data and statistics are shown in Figure 3.
Mean airflow (L/s)
A trend toward an interaction effect between measurement con-
ditions and tube length was found (P = 0.060). Further, there was
a significant main effect of measurement conditions (P < 0.001)
and a nonsignificant main effect of tube length (0.129). Evalu-
ating the main effect of measurement conditions, a significant
decrease was found across pre- and posttreatment measure-
ments (P < 0.001). No significant changes were observed across
the measurements taken for short- and long-duration tasks both
with P values of 1.00. The premeasurements of short- and long-
duration tasks were compared and a nonsignificant difference
of means of 0.020 L/s was found (P = 0.115). The mean AF data
passed the equal variance test (P = 0.635). Power values for data
compiled according to the interaction effect, measurement con-
ditions versus tube length (0.415), and the main effects,
measurement conditions (0.931) and tube length (0.231), were
 ARTICLE IN PRESS
4 Journal of Voice, Vol. ■■, No. ■■, 2017

FIGURE 3. Summary of data and statistical results for the dependent variable: phonation threshold pressure. Notes: Sample size is 20 subjects.
Short and long refer to the duration of the phonatory task. Abbreviations: PTP, phonation threshold pressure.
A. Visual summary of PTP data separated by both tube length and measurement condition. B. Visual summary of PTP data averaged across tube
lengths and separated by measurement condition. C. PTP values collected at each tube length and averaged across tube lengths as well. Note: Values
are represented as mean ± standard deviation. D. Results from two-way repeated measures analysis of variance for PTP. Note: Comparisons with
significant results are denoted with an asterisk. DF, degrees of freedom; SS, sum of squares; MS, mean squares; F, the main or interaction effect
divided by the within variance. E. Results from the post hoc Bonferroni tests. The comparison factor is measurement conditions, with a degree of
freedom of 4. Notes: A negative difference of means was listed for results that decreased from the premeasurement to the postmeasurement of the
condition being compared. Comparisons with significant results are denoted with an asterisk.

evaluated with an alpha of 0.05. Data and statistics are shown
in Figure 4.
Aerodynamic resistance (cmH2O /L/s)
An interaction effect was not found between measurement con-
ditions and tube length (P = 0.172). Further, there was a significant
main effect of measurement conditions (P = 0.021) and a non-
significant main effect of tube length (0.888). No significant
changes were found across pre- and posttreatment (P = 0.137).
Also, no significant changes were observed across the measure-
ments taken for short- and long-duration tasks with respective
P values of 1.000 and 0.255. The premeasurements of short- and
long-duration tasks were compared and a nonsignificant differ-
ence of means of 1.172 cmH2O /L/s was found (P = 1.00). The
AR data passed the equal variance test (P = 0.356). Power values
for data compiled according to the interaction effect, measure-
ment condition versus tube length (0.211), and main effects,
measurement conditions (0.590) and tube length (0.05), were
evaluated with an alpha of 0.05. Data and statistics are shown
in Figure 5.
Contact quotient (%)
An interaction effect between measurement conditions and tube
length was not found (P = 0.489). Further, there was a signifi-
cant main effect of measurement conditions (P = 0.003) and a
nonsignificant main effect of tube length (P = 0.111). Evaluat-
ing the main effect of measurement conditions, no significant
change was found across pre- and posttreatment measure-
 ARTICLE IN PRESS
Randal D Mills, et al Effects of Straw Phonation Through Tubes of Varied Lengths 5

FIGURE 4. Summary of data and statistical results for the dependent variable: mean airflow. Notes: Sample size is 20 subjects. Short and long refer
to the duration of the phonatory task. AF, mean airflow. A. Visual summary of AF data separated by both tube length and measurement condition. B.
Visual summary of AF data averaged across tube lengths and separated by measurement condition. C. AF values collected at each tube length and
averaged across tube lengths as well. Note: Values are represented as mean ± standard deviation. D. Results from two-way repeated measures analysis
of variance for AF. Note: Comparisons with significant results are denoted with an asterisk. DF, degrees of freedom; SS, sum of squares; MS, mean
squares; F, the main or interaction effect divided by the within variance. E. Results from the post hoc Bonferroni tests. The comparison factor is mea-
surement conditions, with a degree of freedom of 4. Notes: A negative difference of means was listed for results that decreased from the premeasurement
to the postmeasurement of the condition being compared. Comparisons with significant results are denoted with an asterisk.

ments, but the results showed a decreasing trend (P = 0.094). No
significant changes were observed across the measurements taken
for short- and long-duration tasks with respective P values of
0.615 and 1.00. The premeasurements of short- and long-
duration tasks were compared and a nonsignificant difference
of means of 1.306% was found (P = 0.434). The CQ data passed
the equal variance test (P = 0.356). Power values for data com-
piled according to the interaction effect, measurement conditions
versus tube length (0.05), and the main effects, measurement con-
ditions (0.846) and tube length (0.260) were evaluated with an
alpha of 0.05. Data and statistics are shown in Figure 6.
Effort and discomfort
Effort and discomfort levels during the short- and long-duration
tasks at each tube length are shown in Figure 7. Participants’
effort and discomfort levels showed an overall increase during
the long-duration task compared with the short-duration task,
while reported levels were similar across tube lengths.
DISCUSSION
Vocalizations following straw phonation exercises have been
shown to result in healthier and more efficient voice produc-
tion. In this study, aerodynamic and EGG measurements were
taken over a range of tube elongations and task durations in
normal-voiced subjects. This study presents and validates a
method to evaluate the alterations of SOVT dimensions and task
durations of straw phonation exercises.
Comparisons across measurement conditions showed a sig-
nificant decrease in oral pressure and AR, a trending decrease
in CQ, and a significant increase on AF. Specifically, these trends
 ARTICLE IN PRESS
6 Journal of Voice, Vol. ■■, No. ■■, 2017

FIGURE 5. Summary of data and statistical results for the dependent variable: aerodynamic resistance. Notes: Sample size is 20 subjects. Short
and long refer to the duration of the phonatory task. AR, aerodynamic resistance. A. Visual summary of AR data separated by both tube length
and measurement condition. B. Visual summary of AR data averaged across tube lengths and separated by measurement condition. C. AR values
collected at each tube length and averaged across tube lengths as well. Note: Values are represented as mean ± standard deviation. D. Results from
two-way repeated measures analysis of variance for AR. Note: Comparisons with significant results are denoted with an asterisk. DF, degrees of
freedom; SS, sum of squares; MS, mean squares; F, the main or interaction effect divided by the within variance. E. Results from the post hoc
Bonferroni tests. The comparison factor is measurement conditions, with a degree of freedom of 4. Notes: A negative difference of means was
listed for results that decreased from the premeasurement to the postmeasurement of the condition being compared. Comparisons with significant
results are denoted with an asterisk.

occurred across pretreatment and posttreatment for all param-
eters. No significant effects were observed for measurement
conditions from pre- to posttask for any parameter. This there-
fore shows that immediate effects of SOVT tasks might not result
in significant changes in voice production. However, as the long-
duration task was always performed after the short-duration task,
it can be stated that the lingering, significant effects of pro-
longed exercises are significant. This assumption was previously
made based on preliminary results and previous literature that
state short-duration task straw exercises may be ineffective.
However, it was verified by finding no statistical change in any
parameters across the short-duration task, both immediately and
15 minutes after as well. Finding lingering effects after the long-
duration task validates the hypothesis that extended duration
SOVT tasks will result in significantly greater changes in pa-
rameters than short-duration tasks. Specifically, increased AF
resulted in an increased supraglottal pressure that caused a greater
push-pull force on the vocal folds and increased inertance, thus
resulting in decreased oral pressure as observed. Decreased CQ
results in less stress to the vocal folds at the same vocal economy,
and lowered AR allows for easier phonation and more inertive
reactance as a component of impedance.
No significant main effect of tube length was found for oral
pressure, AF, AR, and CQ. However, AF and CQ trended toward
significance. This trend signifies that tube lengths could have an
effect on the positive benefits obtained through these exer-
cises. Further, it could be the case that the range of lengths used
was not expansive enough to result in significant differences.
 ARTICLE IN PRESS
Randal D Mills, et al Effects of Straw Phonation Through Tubes of Varied Lengths 7

FIGURE 6. Summary of data and statistical results for the dependent variable: contact quotient. Notes: Sample size is 20 subjects. Short and
long refer to the duration of the phonatory task. CQ, contact quotient. A. Visual summary of CQ data separated by both tube length and measure-
ment condition. B. Visual summary of CQ data averaged across tube lengths and separated by measurement condition. C. CQ values collected at
each tube length and averaged across tube lengths as well. Note: Values are represented as mean ± standard deviation. D. Results from two-way
repeated measures analysis of variance for CQ. Note: Comparisons with significant results are denoted with an asterisk. DF, degrees of freedom;
SS, sum of squares; MS, mean squares; F, the main or interaction effect divided by the within variance. E. Results from the post hoc Bonferroni
tests. The comparison factor is measurement conditions, with a degree of freedom of 4. Notes: A negative difference of means was listed for results
that decreased from the premeasurement to the postmeasurement of the condition being compared. Comparisons with significant results are denoted
with an asterisk.

Future studies should increase the increment levels between SOVT
lengths to evaluate this. It was hypothesized that lengthening the
tube used in SOVT exercises would increase vocal economy. This
is because the effective length of the vocal tract is directly pro-
portional to inertive reactance, which facilitates self-sustained
phonation. Therefore, it would be beneficial to further evaluate
the lengths of SOVTs with this methodology.
The interaction effect of task condition on length resulted in
a significant trend toward the oral pressure and AF (P < 0.1),
These preliminary results strongly suggest that there is an in-
teraction effect of tube length and measurement condition on
voice production. This could imply that various combinations
of tube lengths and task duration could be prescribed for spe-
cific scenarios based on patient needs.
Participants’ discomfort and effort levels showed that task du-
ration had more of an effect than tube length. Specifically,
increasing the task duration increased both levels of discom-
fort and effort. Also, increasing the length of the tube decreased
levels of discomfort and effort as well. Therefore, longer tubes
may be desirable for subjects even if the effects on vocal economy
are not significantly different from shorter tubes, especially for
long-duration tasks.
Although this preliminary study includes high standard de-
viations across human subjects, all standard deviations from
the collected data are within the standard ranges for subjects
aged 18–39 provided by KayPENTAX for PAS measurements.
Within each measurement period, subjects repeated the proto-
col three times to improve accuracy of measurements. Further,
 ARTICLE IN PRESS
8 Journal of Voice, Vol. ■■, No. ■■, 2017
FIGURE 7. Visual summary of questionnaire results averaged separated by both tube length and measurement condition. Note: Short and long
refer to the duration of the phonatory task.
training was completed before measurements to ensure sub-
jects were comfortable with the protocol. This ensured that
subjects were able to consistently produce vocalizations of similar
pitch and loudness.
Future studies using this study design could implement the fol-
lowing improvements. The total subject number could be increased.
An appropriate sample size will increase the power of our statis-
tical tests and ensure less probability of type II error. Before this
study, no literature was available to provide necessary informa-
tion to perform an accurate power analysis. These results provide
expected difference in means, which can benefit future study designs
using a similar methodology. Also, increasing the length incre-
ments of tubes examined could further isolate the effect of tube
length as well as the interaction effect between tube length and
task duration on phonation parameters. Implementation of an in-
traoral tube to obtain supraglottal pressure during voicing through
the tube in the short- and long-duration tasks could allow a more
complete evaluation of the effects of supraglottal pressure on inertive
reactance and phonation threshold pressure. Future studies could
also use KayPENTAX’s CSL module, Analysis of Dysphonia in
Speech and Voice, to obtain acoustic voice quality parameters. Based
on the preliminary results obtained in this study, these modifica-
tions have the potential to produce further significant results.
CONCLUSION
A method focused on altering the effective length of the vocal
tract during straw exercises was used to further study phona-
tion parameters on normal-voiced subjects. It was found that short-
duration tasks do not provide immediate or prolonged effects.
On the contrary, it was found that oral pressure and AR signifi-
cantly decreased, CQ showed a decreasing trend, and AF
significantly increased across long-duration tasks after a period
of 15 minutes of rest. These preliminary results do not show sig-
nificant differences across the tube length used for straw phonation
exercises. However, trends suggest that using this methodolo-
gy, while also increasing the range of lengths evaluated and the
total number of subjects, could provide insight into how alter-
ing SOVT dimensions improves voice production.
Acknowledgments
This study was supported by NIH grant number 2 R01 DC008153
from the National Institute on Deafness and Other Communi-
cation Disorders.
REFERENCES
1. Barkmeier-Kraemer JM, Patel RR. The next 10 years in voice evaluation
and treatment. Semin Speech Lang. 2016;37:158–165.
2. Titze IR. Voice training and therapy with a semi-occluded vocal tract:
rationale and scientific underpinnings. J Speech Lang Hear Res.
2016;49:448–459.
3. Titze IR. Acoustic interpretation of resonant voice. J Voice. 2001;15:519–528.
4. Mannell R. Calculating Resonance Frequencies of Tubes. Macquarie
University; 2008.
5. Titze IR. Vocology. Salt Lake City, UT: National Center for Voice and
Speech; 2012.
6. Guzman M, Castro C, Madrid S, et al. Air pressure and contact quotient
measures during different semioccluded postures in subjects with different
voice conditions. J Voice. 2016;30:759, e1-e10.
7. Story BH, Laukkanen AM, Titze IR. Acoustic impedance of an artificially
lengthened and constricted vocal tract. J Voice. 2000;14:455–469.
8. Master S, Guzman M, Dowdall J. Vocal economy in vocally trained actresses
and untrained female subjects. J Voice. 2013;27:698–704.
 ARTICLE IN PRESS
Randal D Mills, et al Effects of Straw Phonation Through Tubes of Varied Lengths
9. Horacek J, Laukkanen AM, Sidlof P, et al. Comparison of acceleration and
impact stress as possible loading factors in phonation: a computer modeling
study. Folia Phoniatr Logop. 2009;61:137–145.
10. Laukkanen AM, Maki E, Leppanen K. Electroglottogram-based estimation
of vocal economy: “quasi-output-cost ratio.” Folia Phoniatr Logop.
2009;61:316–322.
11. Jiang J, Chen HJ, Stern J, et al. Vocal efficiency measurements in subjects
with vocal polyps and nodules: a preliminary report. Ann Otol Rhinol
Laryngol. 2004;113:277–282.
12. Titze IR, Maxfield L, Palaparthi A. An oral pressure conversion
ratio as a predictor of vocal efficiency. J Voice. 2016;30:398–
406.
View publication stats
9
13. Smith SL, Titze IR. Characterization of flow-resistant tubes used for
semi-occluded vocal tract voice training and therapy. J Voice. 2017;31:113,
e1-e8.
14. Conroy ER, Hennick TM, Awan SN, et al. Effect of variations to a simulated
system of straw phonation therapy on aerodynamic parameters using excised
canine larynges. J Voice. 2014;28:1–6.
15. Kapsner-Smith MR, Hunter EJ, Kirkham K, et al. A randomized controlled
trial of two semi-occluded vocal tract voice therapy protocols. J Speech Lang
Hear Res. 2015;58:535–549.
16. Dargin TC, DeLaunay A, Searl J. Semioccluded vocal tract exercises:
changes in laryngeal and pharyngeal activity during stroboscopy. J Voice.
2016;30:377, e1-e9.