How Stressful Is Deep Bubbling?

Jaana Tyrmi and Anne-Maria Laukkanen, Tampere, Finland

Summary: Objectives. Water resistance therapy by phonating through a tube into the water is used to treat dys-
phonia. Deep submersion (10 cm in water, deep bubbling) is used for hypofunctional voice disorders. Using it with
caution is recommended to avoid vocal overloading. This experimental study aimed to investigate how strenuous deep
bubbling is.
Study Design. Fourteen subjects, half of them with voice training, repeated the syllable [pa:] in comfortable speak-
ing pitch and loudness, loudly, and in strained voice. Thereafter, they phonated a vowel-like sound both in comfortable
loudness and loudly into a glass resonance tube immersed 10 cm into the water.
Methods. Oral pressure, contact quotient (CQ, calculated from electroglottographic signal), and sound pressure level
were studied. The peak oral pressure P(oral) during [p] and shuttering of the outer end of the tube was measured to
estimate the subglottic pressure P(sub) and the mean P(oral) during vowel portions to enable calculation of transglottic
pressure P(trans). Sensations during phonation were reported with an open-ended interview.
Results. P(sub) and P(oral) were higher in deep bubbling and P(trans) lower than in loud syllable phonation, but
the CQ did not differ significantly. Similar results were obtained for the comparison between loud deep bubbling
and strained phonation, although P(sub) did not differ significantly. Most of the subjects reported deep bubbling to
be stressful only for respiratory and lip muscles. No big differences were found between trained and untrained subjects.
Conclusions. The CQ values suggest that deep bubbling may increase vocal fold loading. Further studies should
address impact stress during water resistance exercises.
Key Words: resonance tubeswater resistance therapycontact quotientimpact stressvocal loading.

INTRODUCTION supralaryngeal adjustments that are most beneficial for effi-

Phonation through a tube into the water is used in voice therapy.
The procedure is called water resistance therapy.1 Two types of
tubes are typically used: glass resonance tube (2628 cm in length,
9 mm in inner diameter for adults1) and silicon lax vox tube
(35 cm, 912 mm2). The depth of immersion in water varies from
2 to10 cm (in some cases even 15 cm) for the resonance tube
and from 1 to 7 cm for lax vox tube. A shallow immersion depth
is used for patients with hyperfunctional voice disorder, whereas
deep immersion is used for treatment of hypofunctional voice
disorder or vocal fold pareses.
Water resistance therapy belongs to the so-called semi-
occluded vocal tract exercises (SOVTE), which have been studied
extensively during the last two decades (see eg Refs.310). These
exercises imply a constriction or artificial lengthening of the vocal
tract. Thus, they lower the first vocal tract resonance, which may
increase positive reactance at the fundamental frequency range.4
This may enhance vocal fold vibration.11 Furthermore, when the
constriction of the vocal tract is sufficiently narrow, the airflow
from the vocal tract diminishes, and the oral pressure in-
creases. That, in turn, is prone to relieve adduction mechanically
due to backpressure. Increased oral pressure also intensifies the
vibratory sensations in the vocal tract and facial structures during
phonation. Such sensations help find the laryngeal and
Accepted for publication April 25, 2016.
Results have been presented at the 11th Pan-European Voice Conference PEVOC in
Florence, Italy.
Disclosure: The authors have no conflicts of interest to report.
From the Speech and Voice Research Laboratory, University of Tampere, Tampere, Finland.
Address correspondence and reprint requests to Jaana Tyrmi, Speech and Voice Research
Laboratory, University of Tampere, Virta, kerlundinkatu 5, 33100 Tampere, Finland. E-mail:
jaana.tyrmi@uta.fi
Journal of Voice, Vol. 31, No. 2, pp. 262.e1262.e6
0892-1997
2017 The Voice Foundation. Published by Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.jvoice.2016.04.013
cient and economic voice production.6
Water resistance therapy is a special form of SOVTE as there
the oral pressure can be modified by the depth of immersion of
the tube into the water. The oral pressure needs to be at least
slightly above the pressure of water to start and maintain pho-
nation. Second, during water resistance exercises, the bubbling
of water gives visual and auditory feedback from the airflow rate
used in phonation. Third, the bubbling of water causes oscilla-
tion of oral pressure, which in turn modifies vocal fold vibration
and causes sensations of massage in the vocal tract and the
larynx.12
Former studies on water resistance therapy have mainly focused
on shallow immersion (eg, Refs.1316). Phonation with deep im-
mersion (10 cm in water, deep bubbling) has been regarded
as strenuous, and it is recommended to use it with care and for
short times only.1,17 This study aimed to investigate how stren-
uous deep bubbling is.
Impact stress is regarded as the most important vocal loading
factor.18 It may, thus, be used as a measure of strenuousity of
voice production. According to experimental results, impact stress
increases with subglottic pressure, fundamental frequency, and
adduction of the vocal folds.19 These results are in line with the
clinical and experimental observations, according to which high-
pitched, loud, and pressed phonation leads more easily to
symptoms of vocal fatigue.20,21 Impact stress is very difficult to
measure in humans.22,23 However, contact quotient (CQ) from
electroglottographic (EGG) signal has been found to correlate
with impact stress.24,25 Aerodynamic parameters of voice pro-
duction may also indirectly indicate the amount of vocal loading.
Transglottic pressure P(trans), ie, the difference between sub-
glottic pressure P(sub) and oral pressure P(oral), is the main
driving force of vocal fold vibration (eg,25). When P(trans) in-
creases, it leads to larger vibratory amplitude of the vocal folds
Jaana Tyrmi and Anne-Maria Laukkanen Deep Bubbling
and higher impact stress. Based on these facts, CQ and P(trans)
were used as estimates of strenuousity of voice production in
the present study. P(sub) was estimated from oral pressure during
vocal tract occlusion (eg,26). The CQ and P(trans) from a vowel-
like phonation into resonance tube submerged 10 cm in water
were compared with the same parameters from repetitions of syl-
lable pa produced in comfortable loudness, loudly, and in very
strained (loud, high-pitched, and pressed) voice. Syllable pro-
ductions were used as reference as the vocal tract occlusion during
voiceless plosive p allows for estimating P(sub) from P(oral).26
When phonating into the resonance tube in the water, the vocal
tract occlusion was established by rapid manual shuttering of
the outer end of the tube.5
METHODS
Fourteen subjects with no known pathology of the larynx or voice
(14 subjects; 6 male, 8 female; mean age: 43 years) volun-
teered as participants. The subjects were divided into two groups
according to their prior training in voice. Half of the subjects
(4 female, 2 male) had taken at least their bachelors degree in
vocology (ie, 34 courses in speech training during 3 years of
study), whereas the other half (4 female, 4 male) had not re-
ceived any formal voice training. The subjects with speech training
were familiar with water resistance therapy. The study fol-
lowed the ethical principles of experimental studies in human
subjects.
The subjects phonated (1) three repetitions of [pa:] in their
comfortable conversational speaking pitch and loudness, (2) three
repetitions of the syllable loudly (as when speaking in a large
auditorium without a microphone), and (3) three repetitions of
the syllable in loud, strained voice. Then they produced (4) three
repetitions of vowel-like phonation into a glass resonance tube
immersed 10 cm into water. Comfortable loudness was used. Each
vowel-like tube phonation sample lasted approximately 5 seconds.
Thereafter, (5) the subjects repeated the same tube phonation
task, but this time using loud voice. Loudness was auditorily
monitored by the experimenter during recordings, and sound pres-
sure level was measured from the acoustic recordings. While
phonating into the water (tasks 4 and 5), the outer end of the
tube was rapidly fully closed with a finger at random intervals
(34 times per sample). This shuttering procedure5 allows for
FIGURE 1. (A) The experimental setup for simultaneous registration of acoustic, electroglottographic, and oral pressure signals. (B) The pro-
cedure in shuttering: The outer end of the tube was rapidly fully closed with a finger (top) and then released (bottom picture). This was done at
random intervals during phonation into the tube. The position of the tube in the water was not changed during shuttering.
262.e2
estimating the subglottic pressure from oral pressure during pho-
nation into the tube.
Figure 1 shows the experimental setup. Oral pressure P(oral),
EGG signal, and acoustic signal were recorded simultaneously
in all tasks. P(oral) was registered with MS-110 transducer elec-
tronics unit (Glottal Enterprises, Syracuse, NY) using transducer
PT-75, which is capable of registering also high pressures (up
to 75 cm H2O). The pressure signal was calibrated into cm H2O
values using PC-1H calibration unit. Vocal function was studied
with electroglottography (dual-channel EGG, Glottal Enter-
prises), which shows the glottal contact area variation during
phonation.26 To obtain the EGG signal, two pairs of electrodes
were tightly placed on both sides of the thyroid cartilage using
a Velcro band. The EGG signal was high pass (HP) filtered at
40 Hz, as in Ref.15 The signal quality was monitored on an os-
cillogram during recording. Acoustic signal was recorded with
a head-mounted microphone (AKG C5441, Vienna, Austria) at
6 cm from the lips. The acoustic signal was calibrated for sound
pressure level measurements by recording a sine signal, 1 kHz,
73.6 dB as reference. Recordings were made with Computer-
ized Speech Lab (CSL 4500, KayPENTAX, Lincoln Park, NJ)
using 44.1 kHz sampling rate and 16 bits amplitude quantiza-
tion. Additionally, the subjects were asked on an open-ended
interview about their sensations during the exercise. The inter-
view question was: Describe your sensations during and after
phonation into the tube in the water.
Analyses
Peak P(oral) pressure during [p] in the syllables pa and during
shuttering of the outer end of the tube was measured for an es-
timate of subglottic pressure P(sub). The mean P(oral) during
[a:] in the syllables and during phonation into the water through
the tube while the outer end of it was free (unshuttered) gave
the mean P(oral) during phonation. Transglottic pressure P(trans)
was then calculated as P(sub) P(oral). Analyses were per-
formed with Soundswell Signal Workstation (Nyvalla DSP,
Stockholm, Sweden). The measurement error obtained through
repeated measurements of the same signal was +/ 0.3 cm H2O.
Figure 2 illustrates the pressure measurements.
VoceVista was used to calculate the contact quotient
(CQ = contact time/period time) from EGG (35% threshold level)
262.e3
FIGURE 2. Measurement of the oral pressure signal for (A) syllables and (B) phonation into a tube in the water.
(www.vocevista.com). The signal was visually inspected to ensure
that the automatic period detection was successful. Figure 3 il-
lustrates the CQ calculation.
Statistical treatment
The mean and standard deviation of the parameters were cal-
culated. Significance of differences between sample types was
studied using Students t test or Wilcoxon signed-ranks paired
t test, depending on the population variance. A nonparametric
test was used when the values did not show normal distribu-
tion according to Kolmogorov-Smirnov one-sample t test. The
SPSS-21 software (www.ibm.com) was used in the analyses.
FIGURE 3. Electroglottographic signal. x-axis = time, y-axis = glottal
contact area (increasing contact upwards). Glottal contact quotient, CQ,
calculated as contact time (C, light gray line) divided by period time
(dark gray line). A threshold level of 35% of the signal amplitude was
chosen as the baseline for CQ calculation.27
Journal of Voice, Vol. 31, No. 2, 2017
RESULTS
Table 1 summarizes the results. In deep bubbling at comfort-
able loudness, values for all parameters except for sound pressure
level (SPL) were higher compared with syllable production in
comfortable loudness. P(sub) and P(oral) were also signifi-
cantly higher in comfortable deep bubbling than in loud
phonation, but P(trans) and SPL were significantly lower, and
CQ and P(trans) did not differ significantly. When loud deep
bubbling was compared with strained phonation, no signifi-
cant differences were found for CQ and for P(sub), but P(trans)
and SPL were lower in bubbling.
Figure 4 shows the distribution of CQ and P(trans) values for
the vocally trained and untrained subjects separately. It can be
seen in Figure 4 that there was considerable overlapping in the
parameter values between trained and untrained subjects. The
trained subjects seemed to have a slight tendency for some-
what lower CQ in comfortable and loud syllable repetition and
in deep bubbling in comfortable loudness. Furthermore, the trained
subjects showed somewhat lower P(trans) in loud syllable pro-
duction and in deep bubbling in comfortable loudness.
According to most subjects sensations, deep bubbling
seemed to tire breathing muscles and lip muscles more than the
larynx. Some subjects (one trained and two untrained) felt loud
deep bubbling as strenuous also for the larynx. These sub-
jects, however, did not show different results from the others who
did not report any strenuousity for the larynx.
DISCUSSION
According to the results, CQ did not differ significantly between
loud syllable production and deep bubbling at comfortable loud-
ness. Similarly, CQ values did not differ significantly between
strained syllable production and loud deep bubbling. In both cases,
Jaana Tyrmi and Anne-Maria Laukkanen Deep Bubbling 262.e4

TABLE 1.
Mean Values for CQ, Sub-, Oral-, and Transglottal Air Pressure and Sound Pressure Level (N = 14 Subjects)
CQ (%) P(sub) (cm H2O) P(oral) (cm H2O) P(trans) (cm H2O) SPL (dB)
[pa:] habitual
Mean 44.6 5.2 0.2 5 73.1
SD 8 1 0.1 1 6.1
[pa:] loud
Mean 50.9 9.5 0.3 9.2 83.6
SD 4.9 2.9 0.1 2.9 6
[pa:] strained
Mean 60.1 21.5 0.3 21.3 95.3
SD 7.8 7.8 0.1 7.8 4.7
[u:] tube 10 cm
Mean 54.2 15.7 11.3 6.8 71.6
SD 7.6 4 2.7 3.7 4.9
[u:] tube 10 cm loud
Mean 56 23.4 10.4 13.1 75.9
SD 8.1 6.5 4.2 6 7.8
Differences
[pa:] loud/tube habitual P = 0.103 P = 0.001 (W) P = 0.000 (W) P = 0.011 (W) P = 0.000
[pa:] strained/tube loud P = 0.102 P = 0.484 P = 0.000 P = 0.006 (W) P = 0.000
Notes: Significance of differences between sample types was studied with t test. Because the pressure values for bubbling in comfortable loudness and
P(trans) for loud bubbling did not show a normal distribution, the Wilcoxon signed-rank test was used (W); in other cases Students t test was applied.
Statistically significant P values in bold.
Abbreviations: CQ, contact quotient; SD, standard deviation.

P(sub) was significantly larger in deep bubbling. Relatively high
CQ values and high P(sub) suggest increased vocal effort. Large
CQ may indicate higher impact stress, and thus higher vocal fold
loading.19,25,28 On the basis of the results of the present study, it
seems that deep bubbling may increase vocal fold loading. On
the other hand, because P(oral) was significantly higher in deep
bubbling than in syllable production, P(trans) was also signifi-
cantly lower in bubbling, which seems to suggest a lower vibratory
amplitude of the vocal folds and therefore possibly a lower impact
stress.
There was a slight tendency that untrained subjects had some-
what higher CQ values for comfortable and loud phonation and
for deep bubbling in comfortable loudness, and higher P(trans)
values for loud phonation and for deep bubbling in comfort-
able loudness, compared with the trained subjects (Figure 4).
The trained subjects may, thus, be able to produce loud phonation

FIGURE 4. Distributions for contact quotient (CQ) and P(trans) (N = 14 subjects, 6 with training).
262.e5
and deep bubbling with less effort. Most of them also re-
ported that deep bubbling felt more tiring for respiratory and
lip muscles than for the larynx. Verdolini et al29 also reported
lower CQ values for resonant voice compared with pressed
voice. It may be speculated that the trained subjects most likely
used more resonant kind of voice. On the other hand, opposite
results have also been presented. The classically trained male
singers in the article by Gaskill and Quinney30 showed higher
CQ values than untrained subjects when phonation into a res-
onance tube (the outer end in the air) as well as vowel phonation
pre and post were studied. Similarly, according to Howard et al,31
trained singers had higher CQ values than untrained subjects both
in text reading and in singing. Master et al,32 in turn, did not find
significant difference in CQ values between trained and un-
trained female speakers. The discrepancy in the results may be
related to differences in vocal education and tasks performed.
Deep bubbling may be more strenuous for respiratory and
adductory muscles than for the vocal fold tissue as such. The
EGG waveform modulates during bubbling, which results from
modulation of the vocal fold movement18,33 due to backpressure.
These modulations may reduce the impact stress despite of the
relatively high mean CQ. The mechanical resonance of the vocal
tract may contribute to the oral pressure variation during
bubbling.34 This may intensify the massage-like effect reported.12,15
On the other hand, it may also have negative effects in terms
of potentially increased shear stress.34 Shear stress, however, is
very difficult to study experimentally. The reliability of CQ as
an indicator of impact stress may also be questioned. Accord-
ing to the results of Verdolini et al,23 CQ showed only limited
correlation with impact stress in humans, eg, in pressed phona-
tion CQ was on average lower than in normal phonation, whereas
impact stress was instead much higher in pressed phonation. Sim-
ilarly, the modeling results by Horcek et al33 showed that at high
P(sub) levels the CQ values saturated, whereas the impact stress
values continued to rise. Further studies are needed on the vocal
fold vibratory pattern and true impact stress during bubbling.
The possibility of increased shear stress of the tissue during strong
modulations may also be worth considering in future studies,
eg, by applying modeling.
CONCLUSIONS
Deep bubbling increases vocal effort in terms of P(sub) and
CQ. Therefore, it seems to increase vocal loading. This loading
effect may not be primarily posed on vocal folds, but on respi-
ratory and adductory muscles. Further studies should address
impact stress during water resistance exercises.
REFERENCES
1. Sovijrvi A. Nya metoder vid behandling av rstrubbningar. Nordisk Tidskrift
for Tale og Stemme. 1969;3:121131.
2. Sihvo M. Terve ni, nen Hoidon A B C (Healthy voice. The A B C for
voice care). Helsinki: Kirjapaja; 2006.
3. Laukkanen A-M. About the so called resonance tubes used in Finnish
voice training practice. An electroglottographic and acoustic investigation
on the effects of this method on the voice quality of subjects with normal
voice. Scand J Logop Phoniatr. 1992;17:151161.
4. Story BH, Laukkanen A-M, Titze IR. Acoustic impedance of an artificially
lengthened and constricted vocal tract. J Voice. 2000;14:455469.
Journal of Voice, Vol. 31, No. 2, 2017
5. Titze IR, Finnegan EM, Laukkanen A-M, et al. Raising lung pressure and
pitch in vocal warm-ups: the use of flow-resistant straws. J Sing
2002;58:329338.
6. Titze IR. Voice training and therapy with a semi-occluded vocal tract:
rationale and scientific underpinnings. J Speech Lang Hear Res.
2006;49:448459.
7. Titze IR, Laukkanen A-M. Can vocal economy in phonation be increased
with an artificially lengthened vocal tract? A computer modeling study.
Logoped Phoniatr Vocol. 2007;32:147156.
8. Laukkanen A-M, Titze IR, Hoffman H, et al. Effects of a semi-occluded
vocal tract on laryngeal muscle activity and glottal adduction in a single
female subject. Folia Phoniatr Logop. 2008;60:298311.
9. Vampola T, Laukkanen A-M, Horcek J, et al. Vocal tract changes caused
by phonation into a tube: a case study using computer tomography and
finite-element modeling. J Acoust Soc Am. 2011;129:310315.
10. Guzman M, Laukkanen A-M, Krupa P, et al. Vocal tract and glottal function
during and after vocal exercising with resonance tube and straw. J Voice.
2013;27:523.e19523.e34.
11. Titze IR, Story BH. Acoustic interactions of the voice source with the lower
vocal tract. J Acoust Soc Am. 1997;101:22342243.
12. Enflo L, Sundberg J, Romedahl C, et al. Effects on vocal fold collision and
phonation threshold pressure of resonance tube phonation with tube end in
water. J Speech Lang Hear Res. 2013;56:15301538.
13. Wistbacka G, Sundberg J, Simberg S. Vertical laryngeal position and oral
pressure variations during resonance tube phonation in water and in air. A
pilot study. Logoped Phoniatr Vocol. 2015;17. doi:10.3109/14015439
.2015.1028101.
14. Amarante Andrade P, Wistbacka G, Larsson H, et al. The flow and pressure
relationships in different tubes commonly used for semi-occluded vocal tract
exercises. J Voice. 2016;1:3641. doi:10.1016/j.jvoice.2015.02.004.
15. Granqvist S, Simberg S, Hertegrd S, et al. Resonance tube phonation in
water: high-speed imaging, electroglottographic and oral pressure
observations of vocal fold vibrationsa pilot study. Logoped Phoniatr Vocol.
2015;3:113121. doi:10.3109/14015439.2014.913682.
16. Paes S, Zambon F, Yamasaki R, et al. Immediate effects of the Finnish
resonance tube method on behavioral dysphonia. J Voice. 2013;27:717
722.
17. Simberg S, Laine A. The resonance tube method in voice therapy: description
and practical implementations. Logoped Phoniatr Vocol. 2007;32:165170.
18. Titze IR. Mechanical stress in phonation. J Voice. 1994;8:99105.
19. Jiang JJ, Titze IR. Measurement of vocal fold intraglottal pressure and impact
stress. J Voice. 1994;8:132144.
20. Rantala L, Vilkman E. Relationship between subjective voice complaints
and acoustic parameters in female teachers voices. J Voice. 1999;13:484
495.
21. Mki E, Niemi HM, Lundn S, et al. F0, SPL and vocal fatigue in a vocally
loading test. Proc 25th World Congr Int Assoc Logop Phoniatr, Montreal;
CDROM, 2001.
22. Reed C, Doherty E, Shipp T. Direct measurement of vocal fold medial forces.
Am Speech Hear Ass Rep. 1992;34:131 (A).
23. Verdolini K, Hess MM, Titze IR, et al. Investigation of vocal fold impact
stress in human subjects. J Voice. 1999;13:184202.
24. Verdolini K, Chan R, Titze IR, et al. Correspondence of electroglottographic
closed quotient to vocal fold impact stress in excised canine larynges. J Voice.
1998;12:415423.
25. Titze IR. Principles of Voice Production. 2nd ed. Iowa: NCVS; 2000.
26. Baken RJ, Orlikoff RF. Clinical Measurement of Speech and Voice. San
Diego, CA: Singular/Thomson Delmar; 2000.
27. Rothenberg M, Mashie JJ. Monitoring vocal fold abduction through vocal
fold contact area. J Speech Hear Res. 1988;31:338351.
28. Horcek J, Laukkanen A-M, idlof P, et al. Comparison of acceleration and
impact stress as possible loading factors in phonation. A computer modeling
study. Folia Phoniatr Logop. 2009;61:137145.
29. Verdolini K, Druker DG, Palmer PM, et al. Laryngeal adduction in resonant
voice. J Voice. 1998;12:315327.
30. Gaskill CS, Quinney DM. The effect of resonance tubes on glottal contact
quotient with and without task instruction: a comparison of trained and
untrained voices. J Voice. 2012;26:e79e93.
Jaana Tyrmi and Anne-Maria Laukkanen Deep Bubbling
31. Howard DM, Lindsey GA, Allen B. Toward the quantification of vocal
efficiency. J Voice. 1990;4:205212.
32. Master S, Guzman M, Carlos de Miranda H, et al. Electroglottographic
analysis of actresses and nonactresses voices in different levels of intensity.
J Voice. 2013;27:187194.
33. Horcek J, Laukkanen A-M, vec JG. Closed quotient as an estimate of
impact stress: A computer modelling study. Proceedings of the 7th
262.e6
International Conference Advances in Quantitative Laryngology, Voice and
Speech Research. Groningen; 2006.
34. Horcek J, Laukkanen A-M, Radolf V. Low frequency mechanical resonance
of the vocal tract in water resistance therapy. In: Manfredi TC, ed.
Proceedings of the 9th International Workshop Models and Analysis of
Vocal Emissions for Biomedical Applications (MAVEBA). Firenze, Italy:
Firenze University Press; 2015:143145.