Logopedics Phoniatrics Vocology.

2007; 32: 147
156

ORIGINAL ARTICLE

Can vocal economy in phonation be increased with an artificially

lengthened vocal tract? A computer modeling study

INGO R. TITZE1,2 & ANNE-MARIA LAUKKANEN3

1

Department of Speech Pathology and Audiology, The University of Iowa, Iowa City, IA, USA, 2National Center for Voice
and Speech, The Denver Center for the Performing Arts, Denver, CO, USA, 3Department of Speech Communication and
Voice Research, University of Tampere, Tampere, Finland

Abstract
Voiced obstruents and phonation into tubes are widely used as vocal exercises. They increase the inertive reactance of the
vocal tract in the 200
1000 Hz range and thereby reinforce vocal fold vibration. But the effect is strong only when the
epilarynx tube is also narrowed. The present study focused on the effects of a resonance tube (27 cm in length, 0.5 cm2
cross-sectional area, hard walls) on vocal tract reactance and the accompanying economy of voice production (defined as
maximum flow declination rate (MFDR), divided by maximum area declination rate (MADR)). The vowel /u/ and
phonation into the tube were simulated with a computer model. Three values were given to the cross-sectional area of the
epilarynx tube (0.2 cm2, 0.5 cm2, and 1.6 cm2), which is at the opposite end of the vocal tract from the artificial resonance
tube. The degree of glottal adduction was varied in order to find the economy maximum for each epilarynx tube setting.
Results showed that the resonance tube lowered F1 from 300 Hz to 150 Hz and doubled the vocal tract inertive
reactance at F0 100 Hz. The largest economy with the resonance tube was obtained when the epilarynx tube was
narrowed (relative to the rest of the vocal tract) and sufficiently tight adduction was used. Most importantly, the intraoral
acoustic pressure (calculated at 0.8 cm behind the lips) was tripled with the tube. The results suggest that by optimizing the
vibratory sensations in the face that are attributed to increased intraoral acoustic pressure, phonation into a tube may assist a
trainee in finding an optimal glottal and epilaryngeal setting for the greatest vocal economy.

Key words: Airflow, breath control, computer modeling, epilaryngeal narrowing, vocal economy, voice training and therapy

Introduction
Voiced fricatives like /v, z, b/, lip and tongue trills,
nasal consonants, and phonation into tubes have
been widely used in voice training and therapy (1
5).
Beneficial effects have also been reported when a
person phonates against a hand nearly covering
the mouth (6). Lessac (7) has proposed the use
of a y-buzz as a vocal exercise, which is a closed
front vowel produced with a slight protrusion of
the lips and with so narrow a constriction between
the tongue and the palate that it almost sounds
like the semivowel /j/. This y-buzz exercise and
other components of a series of energy principles
described by Lessac have been crafted into the
Lessac-Madsen Resonant Voice Therapy method
by Verdolini (8). Also, Stemples Vocal Function

Exercises (9) utilize the vowel /o/ as a primary vocal

tract configuration for practice. Collectively, we refer
to all of these exercises as semiocclusive vocal tract
exercises (10).
Some authors have suggested that exercises on
voiced fricatives also increase breath management in
singing (3) and for general improvement of breathing (11,12). Phonation into tubes has been used in
speech therapy for the treatment of hypernasality
(4,13,14), in voice therapy for the treatment of both
phonasthenia and hyperfunctional voice disorder
(4,13
18), and in voice training to improve voice
quality and projection (19).
Phonation into glass tubes (25
28 cm in length, 8

9 mm inner diameter), called resonance tubes, has
been used in Finnish voice training and therapy
practice (5,13,15
18,20) and in Norway (21). In

Correspondence: Ingo R. Titze PhD, National Center for Voice and Speech, The Denver Center for the Performing Arts, 1101 13th Street, Denver, CO
80204, USA. Fax: 1-303-893-6487. E-mail: ititze@dcpa.org

(Received 2 February 2006; accepted 9 February 2007)

ISSN 1401-5439 print/ISSN 1651-2022 online # 2007 Taylor & Francis
DOI: 10.1080/14015430701439765

148

I. R. Titze & A.-M. Laukkanen

Germany, Gundermann (14) and Habermann (4)

mention a method of humming on /m, n, l/ into a
glass tube (B12 cm in length, 1 cm inner diameter)
proposed by Spiess (22) and later recommended by
Stein (23). The name resonance tube comes from
the strong sensations of vibrations that are felt in the
lips and face during phonation into these tubes. A
resonance tube is used either so that one end of it is
sunk into a cup filled with water (water resistance
therapy, see e.g. Sovijarvi (20), Rauhala (15)) or so
that it is free in the air, pointing straight out of the
subjects mouth as a natural extension of the vocal
tract. The other end is kept firmly between the lips.
The subjects are instructed to produce a vowel-like
sound (/u, y/ are the most natural choices) into the
tube. The aim is the most comfortable, effortless
phonation that produces maximum vibratory sensations in the lips and face. According to subjective
sensations of many trainees, phonation feels easier
and the voice sounds louder immediately after
exercising with the tubes (clinical observation by
author A.-M. Laukkanen, who has administered the
therapy often). According to Tapani (16) and Simberg (18), patients suffering from functional and
other voice disorders seem to have derived benefit
from the therapy.
Some studies of the instantaneous effects of vocal
tract occlusions on vocal fold vibration have been
carried out on human subjects. Bickley and Stevens
(24), using acoustic analysis in combination with
electroglottography, reported an increase in the open
quotient and a steeper spectral slope in the glottal
source function as a consequence of vocal tract
constriction. These results have recently been confirmed with a computer model (10). However,
Laukkanen (19,25) obtained opposite results with
electroglottography (EGG) during and immediately
after phonation on /b/ and into resonance tubes with
subjects that had received training in the use of the
semiocclusive. The relative open time was reduced
during and after semiocclusion with a bilabial
fricative and a tube. Also, the average laryngeal
muscle activity was the same or lower during
phonation into a resonance tube or on /b/ compared
to vowel phonation (26,27). Decreased glottal resistance due to increased flow has been observed
immediately after 1 minute exercising on /b, m/ and
the resonance tubes (28,29). These studies suggest
that vocalists can learn to compensate for the
semiocclusion and perhaps utilize it to their advantage in training the vocal fold adduction and the
related musculature.
All the above-mentioned exercises imply that
semiocclusion of the vocal tract (steady or time
varying as in a lip trill) or an artificial lengthening of
the vocal tract increase the source-vocal tract inter-

actions. Modeling studies have shown that vocal

tract input impedance (and particularly the inertive
reactance) increases with a tube that lengthens the
vocal tract (30), and the glottal flow amplitude and
pulse shape change with increased inertive reactance
(31
33). Furthermore, oscillation threshold pressure
is reduced by increased vocal tract inertance (34). A
study with a singer also suggested an effect of vocal
tract inertance on the oscillatory characteristics of
the vocal folds (35).
Inertive reactance in the vocal tract can also be
increased by narrowing the epilarynx tube area
instead of semioccluding the mouth (36). This also
lowers phonatory threshold pressure and increases
maximum flow declination rate (33), leading to
strengthening of the higher harmonics and even to
an increase in sound pressure level (SPL). This
increase in maximum flow declination rate (MFDR)
would imply a more economic voice production
(more sound output with less mechanical stress
imposed on the vocal fold tissue), provided that the
maximum area declination rate (MADR) in the
glottis does not increase proportionately. We have
suggested the use of the ratio MFDR/MADR as a
measure of vocal economy (33).
Phonation into narrow straws, as opposed to
longer and wider tubes, may add another benefit.
High subglottic pressures are possible without excessive collision of the vocal folds. Titze et al. (37)
observed lower amplitude and a lower relative closed
time of the glottis on an EGG signal when phonation
into straws was compared to vowel phonation. It was
concluded that, with narrow straws, it is possible to
exercise the use of the high subglottic pressures
needed in singing, while having minimal collision of
the vocal folds. During phonation into a narrow
straw, the intraglottal air pressure rises, causing the
vocal folds to abduct, thereby diminishing the
collision force during voice production.
The current study focused on the use of one
specific resonance tube used in Finland. First,
the effects of the tube on vocal tract reactance were
calculated. Second, the effects of the tube on selfsustained vocal fold oscillation were studied with a
computer simulation model.
Methods
The three-mass body-cover model of Story and
Titze (38) was used for simulation. The model
allowed inputs in the form of laryngeal muscle
activation (39). The vocal tract was simulated with
the wave reflection algorithm (40,41), including
frictional air losses, kinetic losses, wall vibration
losses, radiation losses, and glottal losses. The
supraglottal tract was modeled with 44 sections,

Vocal economy study with artificially lengthened tract 149

each 0.398 cm in length, and cross-sections for the
/u/ vowel determined experimentally with magnetic
resonance imaging by Story et al. (42). The total
length of the supraglottal vocal tract was 17.5 cm,
which corresponds to an average male vocal tract. A
subglottal tract (36 sections, 14 cm in length) was
included, with the area function also modeled after
Story et al. (42).
Sound radiation from the lips was modeled as a
circular piston oscillating in a spherical baffle, which
has become a standard in speech simulation (43), but
can be challenged for frequencies above 5000 Hz.
In fact, many aspects of the wave reflection algorithm
as detailed by Liljencrants (40) and Story (41) begin
to lose accuracy for frequencies above 5000 Hz
because they are based on a one-dimensional wave
equation.
Waveforms were simulated with this model, typically 200 ms in length to show about 20 cycles of
vibration at around 100 Hz. From these waveforms,
the following variables were calculated: peak glottal
area, mean glottal area, MADR, peak glottal flow,
mean glottal flow, MFDR, vocal economy (MFDR/
MADR), and glottal efficiency (acoustic output
power divided by the product of mean airflow and
subglottic pressure) (44). In addition, several values
of peak and mean vocal tract pressures were computed. Results are shown in Table I. Finally, the
combined reactance of the subglottal and supraglottal vocal tract was calculated with and without the
tube, and with three epilarynx tube diameters. The
reactance calculations followed the procedure described by Story et al. (30).
It must be pointed out that the accuracy of any
simulation depends on many factors. Some parameters in the model are known to better than 0.1%
accuracy (e.g. density of air, sound velocity), but
other parameters are known only to an order-ofmagnitude (e.g. tissue viscosities and elasticities).
Thus, the results that are about to be shown may
have error in an absolute sense, but the relative
changes with parameter variation, which are of
primary importance, are less susceptible to error
because the uncertainties usually cancel out.
Results
The vowel /u/ was first simulated as a control case.
The vocal tract shape is shown in Figure 1 (top left).
This vowel has a small lip opening, making the
radiation losses comparable to those of the tube.
Simulated laryngeal muscle activity (20% thyroarytenoid, 20% cricothyroid, and 50% lateral cricoarytenoid) produced an F0 of about 100 Hz. The
epilarynx tube cross-sectional area was 0.5 cm2 (the
first eight supraglottal sections), and the lung

pressure was 0.8 kilo-pascals (kPa). The value 0.5

cm for Ae is typical on the basis of measurements
made by Story et al. (45). Several studies of how
vocal efficiency and vocal economy vary with Ae
have already been conducted (46
48). Results follow
the basic principles of maximum power transfer in
electrical and acoustic circuits. If Ae is such that the
vocal tract input impedance matches the glottal
impedance (which is a time-varying nonlinear quantity), the output power of the simulator is maximized. Efficiency of conversion of aerodynamic
power to acoustic power, on the other hand, is not
maximized when the impedances match. As an
alternative to glottal efficiency, we have been attracted to a quantity called vocal economy, which is
presently defined as the ratio of maximum flow
declination rate to maximum area declination rate
(10). In the current study, the value of Ae allowed
vocal economy to reach a peak with various glottal
adjustments. Values of Ae B0.1 prevented vocal fold
oscillation because the input impedance was too high
and values of Ae 2.0 greatly reduced the oscillation
range because no benefit was obtained from vocal
tract coupling.
Returning to Figure 1, the following output waveforms of the model are shown on the left panel, top
to bottom: contact area (ca) of the vocal folds in
cm2, glottal area (ga) between the vocal folds in cm2,
glottal airflow (ug) in L/s, and glottal flow derivative
(dug) in m3/s2. On the right panel, we see vocal tract
pressures in kPa from top to bottom: lip-radiated
output pressure (Po), intraoral (mouth) pressure
(Pm) at a location 0.8 cm behind the lips, epilarynx
tube input pressure (Pe), intraglottal pressure (Pg),
and subglottic pressure (Ps). To observe their
relative sizes, all pressures are scaled equally between
2.0 and 2.0 kPa. Note the relatively small lipradiated pressure (top right) in comparison to the
pressures below, within, and above the glottis
(bottom three). The intraoral pressure (second
from top) is also relatively small for the vowel /u/.

Effects of vocal tract lengthening with a tube

Figure 2 shows the same set of simulated waveforms
when a resonance tube is added at the lips, 27 cm in
length and 0.5 cm2 in cross sectional area, the same
as the epilarynx tube. With ordinary speech airflows,
little air turbulence was noted when a subject
phonated through this tube. Hence, no turbulence
was simulated with noise sources. The most outstanding visible feature in Figure 2 is the large
intraoral acoustic pressure (second from top on the
right). This pressure (Pm) is increased by a factor of
three over the vowel /u/ without a tube. We believe
that this large mouth pressure can be felt as a

150

Vocal tract
configuration

Peak area Mean area

(cm2)
(cm2)

/u/; Ae 0.5 cm2;

0.67
LCA 50%
Resonance tube Ae 1.6 cm2
46% LCA
0.41
47%
0.42
48%
0.40
49%
0.54
50%
0.34
51%
0.48
52%
0.44
53%
0.19
Ae0.5 cm2
46% LCA
47%
48%
49%
50%
51%
52%
53%
Ae0.2 cm2
51% LCA
52%
53%
54%
55%
(56% does not
phonate)

MADR
(cm2/ms)

Peak flow Mean flow

(L/s)
(L/s)

MFDR
(cm3/s2)

Peak Pg
(kPa)

Mean Pg
(kPa)

Peak Pe
(kPa)

Mean Pe
(kPa)

Peak Pm
(kPa)

Mean Pm
(kPa)

Economy
(cm/ms)

Efficiency

0.23

0.45

0.87

0.33

4.12

4.80

0.48

3.27

0.29

0.48

0.017

9.09

0.0003150

0.24
0.23
0.20
0.20
0.15
0.17
0.14
0.057

0.09
0.17
0.12
0.29
0.11
0.30
0.29
0.12

0.66
0.64
0.61
0.68
0.61
0.67
0.65
0.43

0.51
0.43
0.38
0.29
0.31
0.24
0.22
0.15

0.23
0.30
0.44
1.65
0.61
1.73
1.69
0.57

0.99
1.05
1.08
1.02
1.19
1.10
1.06
1.04

0.53
0.58
0.59
0.39
0.68
0.38
0.38
0.61

0.59
0.62
0.69
1.02
0.78
1.14
1.11
0.63

0.38
0.33
0.30
0.22
0.24
0.18
0.16
0.11

0.32
0.55
0.70
0.92
0.75
1.04
1.02
0.60

0.080
0.073
0.059
0.040
0.041
0.032
0.030
0.021

2.56
2.80
3.85
5.66
5.51
5.77
5.75
4.70

0.0000001
0.0000027
0.0000057
0.0000272
0.0000119
0.0000471
0.0000426
0.0000158

0.43
0.44
0.43
0.61
0.63
0.56
0.26
0.18

0.24
0.24
0.21
0.22
0.21
0.19
0.10
0.053

0.10
0.12
0.14
0.37
0.36
0.38
0.13
0.12

0.59
0.58
0.56
0.56
0.72
0.57
0.49
0.41

0.47
0.39
0.33
0.26
0.29
0.21
0.21
0.14

0.14
0.44
0.54
2.53
2.76
2.77
0.73
0.61

1.07
1.19
1.14
2.74
3.20
3.00
1.16
1.12

0.55
0.61
0.65
0.42
0.51
0.39
0.60
0.63

0.63
0.74
0.82
1.80
2.17
2.04
1.05
0.82

0.44
0.39
0.32
0.25
0.26
0.20
0.20
0.13

0.30
0.70
0.81
0.99
1.33
1.12
0.85
0.69

0.071
0.057
0.052
0.041
0.029
0.028
0.032
0.019

1.48
3.80
3.94
6.93
7.65
7.29
5.67
5.17

0.0000007
0.0000064
0.0000099
0.0000484
0.0000844
0.0001179
0.0000250
0.0000251

0.51
0.46
0.45
0.43
0.45

0.15
0.12
0.12
0.08
0.12

0.31
0.30
0.35
0.31
0.37

0.33
0.33
0.33
0.33
0.33

0.16
0.14
0.13
0.11
0.11

2.61
2.59
3.09
2.57
3.27

5.00
5.24
5.44
4.91
5.60

0.59
0.46
0.47
0.51
0.43

4.13
4.02
4.30
3.79
4.43

0.25
0.22
0.20
0.18
0.17

0.78
0.75
0.78
0.76
0.74

0.022
0.016
0.014
0.017
0.015

8.53
8.76
8.95
8.28
8.92

0.0000933
0.0001388
0.0001497
0.0000971
0.0002841

Notes: MADRmaximum area declination rate; MFDRmaximum flow delination rate; LCA lateral cricoarytenoid activity; SPLsound pressure level; Pg pressure in the glottis; Pm
pressure in the mouth (behind the lips); Pe pressure at the epilarynx tube entry.

I. R. Titze & A.-M. Laukkanen

Table I. Results for simulations of /u/ in the top row, and with a resonance tube in the remaining rows. For each of three cross-sectional areas of the epilarynx (Ae) there are several degrees of
adduction. Vocal economy is defined as (MFDR/MADR) and efficiency as (SPL/mean flow  mean subglottic pressure). In bold: Values for the degree of adduction (in % LCA) giving the highest
economy.

Vocal economy study with artificially lengthened tract 151

Figure 1. Example of some outputs of the model (vowel /u/, 50% simulated lateral cricoarytenoid (LCA) adduction, 0.5 cm2 epilarynx
tube). Left column from top: Schematic picture of the cross-sectional area of the trachea, glottis, epilarynx tube and mouth cavity; vocal fold
contact area (ca); glottal area (ga), glottal airflow (ug); first derivative of glottal flow (fug, negative peak shows the maximum flow
declination rate). Right column from top: oral radiated air pressure (Po); mouth pressure 0.8 cm behind lips (Pm); epilarynx tube input
pressure (Pe); intraglottal pressure (Pg); subglottic pressure (Ps).

buzzing in the lips and other facial tissues when a

person phonates into a tube.

Effects of combined epilarynx tube narrowing with vocal

tract lengthening on vocal tract reactance
Figure 3(a) shows the shapes for the vowel /u/ again
without a resonance tube, but this time with three
different values of epilarynx tube cross-section from
top to bottom: 1.6 cm2, 0.5 cm2, and 0.2 cm2.
Figure 3(b) shows the corresponding reactance
curves of the vocal tract shapes. Reactance is
expressed in units of dyn-s/cm5, where 1 dyn-s/
cm5 105 Pa-s/m3. Thin solid lines are for supraglottal reactance, dashed lines for subglottal reactance, and thick solid lines for the combined
reactance. It can be seen that narrowing of the

epilarynx tube area from 1.6 cm2 to 0.2 cm2 (top to

bottom) approximately doubled the reactance at
frequencies below 300 Hz (e.g. from 10 dyn-s/cm5
at 100 Hz to 20 dyn-s/cm5 at the same frequency).
This increased reactance gives rise to greater reinforcement of vocal fold vibration due to delayed
feedback from this reactive load (34).
Figure 4 shows similar results when the resonance
tube is added to the vocal tract. The tube lowered F1
from about 300 Hz to 150 Hz. This further
increased the positive (inertive) reactance below
F1. For example, 100 Hz is increased from 20 dyns/cm5 to 40 dyn-s/cm5 (2106 to 4 106 Pa-s/m3).
But negative reactance occurred from 150 Hz to
about 250 Hz. This is an area where vocal fold
vibration is not enhanced by the vocal tract. The
region of negative reactance can be shrunk, however,

152

I. R. Titze & A.-M. Laukkanen

Figure 2. Outputs of the model for the vowel /u/ with a 27-cm tube attached (top left). All waveforms are comparable to those of Figure 1.

by narrowing the epilarynx tube, as is shown in the

lower panels of Figure 4. Due to second formant
lowering with the tube, positive reactance also
increased in the 400
600 Hz region. This effect on
higher (singing) fundamental frequencies and their
harmonics will be left as a follow-on study. Here we
are concerned only with reactive effects at normal
speaking fundamental frequencies.

Effects of epilaryngeal narrowing and the resonance tube

on vocal economy in voice production
As stated earlier, vocal economy is still in the process
of being developed. Our current definition is
MFDR/MADR, based on a simple glottal geometry
that does not include anterio-posterior variation
(33). MFDR is the maximum flow declination rate
and MADR is the maximum area declination rate.
As it presently stands, the ratio MFDR/MADR has
dimensions of velocity (m/s), which has no strong

physical interpretation. It does, however, relate

abruptness of airflow change to abruptness of tissue
velocity change, the first being desirable for acoustic
excitation and the second being undesirable for
tissue stress. Thus, the higher the ratio is, the greater
the economy of production (in theory). As more
sophisticated vocal fold models are used, threedimensional glottal kinematics may be needed to
refine the definition. For the present investigation,
the definition is adequate.
It has been shown that vocal economy (however
defined) is likely to be a function of vocal fold
adduction (10,49). Hence, a third experimental
variable, vocal fold adduction, was included in the
simulation. For each value of epilarynx tube area, as
well as for the tube versus no-tube condition, a
group of values for simulated lateral cricothyroid
(LCA) muscle activity was chosen to find the
optimum value of adduction. The highest value of
vocal economy was the function to be optimized.

Vocal economy study with artificially lengthened tract 153

Figure 3. (a) Vocal tract shape for the /u/ vowel and (b) with the corresponding reactance curve for three epilaryngeal settings: 1.6 cm2
(top), 0.5 cm2 (middle), 0.2 cm2 (bottom). Thin solid linesubglottic reactance, dashed linesupraglottic reactance, thick linetotal
vocal tract reactance (1 dyn-s/cm5 105 Pa-s/m3).

Table I summarizes the results obtained for

selected variables calculated from the waveforms.
These variables are labeled across the top. The rows
are divided into four groups, the /u/ vowel being in
row 1 as a control case (with 50% LCA and 0.5 cm2
epilarynx tube), followed by three groups of resonance tube cases for different epilarynx tube crosssections Ae. Each group of Ae contains several
simulated LCA activities. Simulated LCA activity
was varied such that a peak value in vocal economy
was established, with values dropping off on either
side. The row with bold numbers shows the maximum economy case. Note that for Ae 1.6 cm2, the
peak economy value is 5.77 cm/s, while for Ae 
0.5 cm2 it is 9.65 cm/s, and for Ae 0.2 cm2 it is
8.95 cm/s. These optimized economy cases yield the
primary numbers for comparison. When the optimum economy cases with the resonance tube are
compared to the /u/ vowel, both peak and mean

glottal areas (first two columns) and glottal flows

(fourth and fifth columns) generally declined
slightly, suggesting greater steady back pressures on
the vocal folds and smaller vibrational amplitudes
when the resonance tube is attached. MADR and
MFDR are also lower with the resonance tube.
Acoustic pressures along the vocal tract (Pg 
intraglottal pressure, Pe  epilaryngeal tube pressure, Pm mouth pressure behind the lips) are
likewise generally lower with the resonance tube.
There is one major exception: Pm. This mouth
pressure behind the lips increased dramatically for all
cases with the resonance tube, which is perhaps the
most significant result of this study.
The highest economy value with the resonance
tube (8.95 cm/s) was obtained with the narrowest
epilarynx. It was very close to the value for /u/, 9.09
cm/s. Efficiency is more difficult to compare because
the tube radiates energy differently than the lips.

154

I. R. Titze & A.-M. Laukkanen

Figure 4. (a) Vocal tract shape for the /u/ vowel combined with a resonance tube and (b) the corresponding reactance curve for three
epilaryngeal settings: 1.6 cm2 (top), 0.5 cm2 (middle), 0.2 cm2 (bottom). Thin solid linesubglottic reactance, dashed linesupraglottic
reactance, thick linetotal vocal tract reactance (1 dyn-s/cm5 105 Pa-s/m3).

Vocal efficiency, traditionally defined as the ratio of

radiated power from the mouth to aerodynamic
power at the glottis, has limited use because it is so
highly dependent on mouth opening. Every vowel
has a different efficiency. Vocal economy, as defined
here, is less sensitive to vowel because the computation involves glottal variables only.
Discussion
It is known on the basis of earlier results (33) that a
relatively narrow laryngeal vestibule (epilarynx tube)
can increase the maximum flow declination rate
(MFDR) while simultaneously lowering the mean
glottal airflow. Since a narrowed epilarynx tube
causes some steady backpressure in the glottis, it
also diminishes the maximum area declination rate
(MADR) and thus leads to higher economy. This
increase in vocal economy can be linked to an

increase in vocal tract inertive reactance, which

assists the vocal folds in self-sustained oscillation.
In this study, a resonance tube added to the vocal
tract at the lips in and of itself increased the inertive
reactance in the 100
200 Hz region, which could
then be further increased if the epilarynx tube was
also narrowed. But the economy was not greater
than that of an /u/ vowel, which has a lip opening
comparable to the tube diameter (between 0.2 and
0.5 cm2). Thus, the tube seemed to offer no more
than any other oral semiocclusive. In particular,
there was no new resonance at speaking pitches.
Remarkable, however, was the finding that the
mouth pressure just behind the lips was three times
higher with the tube than with an /u/ vowel. Therefore, it seems plausible that the rationale of using a
tube with vocal exercising is that it guides the trainee
to the sensation of facial tissue vibration, which is
sensitive to impedance matching between the glottis,

Vocal economy study with artificially lengthened tract 155

epilarynx, and the vocal tract. It is likely that the
most beneficial epilaryngeal setting cannot easily be
found without this lip buzz amplifier.
Regarding the concept of narrowing the epilarynx
tube, a note of caution is offered. If not carefully
conceptualized and executed, epilarynx tube narrowing may be interpreted as hyper-adduction of the
false folds. The opposite is true. Narrowing of the
epilarynx tube should take place only by anterioposterior movement of the epiglottis, not by mediolateral movement of the ventricular folds, which
could easily be set into vibration. This vibration
would be rough, with a strained voice quality.
Traditionally, voice coaches and singing pedagogues
have stressed the importance of a wide pharynx (e.g.
Appelman) (50). The mental image of a yawn (or, at
least, an anticipation of a yawn) is promoted as a
means of freeing up the voice. Widening the pharynx
(and perhaps the entire vocal tract) effectively
narrows the epilarynx tube, if held constant. Acoustically, narrowing and widening are relative concepts.
What matters from the point of view of impedance
matching is the relative size between the mean crosssectional area at the entry of the glottis and the crosssectional area of the vocal tract. This can be obtained
in a variety of ways.
Conclusion
Phonation into a so-called resonance tube, although
not providing any new resonance conditions other
than what is predicted from an artificially lengthened
vocal tract, appears to have therapeutic value in that
it provides acoustic pressure feedback from the lip
area. Relatively strong pressures are felt at the liptube junction, which increase when the epilarynx
tube area above the glottis effectively narrows. Thus,
as has been claimed in earlier studies (10), altering
the acoustic load at the mouth with a tube may
facilitate a better impedance match at the glottis.
It remains to be shown whether specific lengthdiameter combinations of the tube can optimize the
process of impedance matching. Future studies will
focus on the laryngeal and epilaryngeal settings of
human subjects during and immediately after vocal
exercising with tubes and other occlusions of the
vocal tract.

Acknowledgements
This study was supported by funding from the
National Institute on Deafness and Other Communication Disorders, grant number 1R01 DC04347,
and grant numbers 32879 and 106139 from the
Academy of Finland.

References
1. Berry C. Your voice and how to use it successfully. London:
Harrap; 1975.
2. Linklater K. Freeing the natural voice. New York: Drama
Book Specialists; 1976.
3. Carroll LM, Sataloff RT. The singing voice. In: Sataloff RT,
editor. Professional Voice: The Science and Art of Clinical
Care. New York: Raven Press Ltd; 1991.
4. Habermann G. Funktionelle Stimmstorungen und ihre Behandlung. Arch Otorhinolaryngol. 1980;227:171
345.
anifysiologiasta ja artikulaatiotekniikasta [On
5. Sovijarvi A. A
voice physiology and articulatory technique]. Department of
Phonetics, University of Helsinki; 1966. (In Finnish)
6. Aderhold E. Sprecherziehung des Schauspielers. Grundlagen
und Methoden. Berlin: Henschelverlag; 1963.
7. Lessac A. The use and training of the human voice. A
practical approach to voice and speech dynamics, 2nd ed.
New York: DBS Publications; 1967.
8. Verdolini K. Resonant voice therapy. In: Stemple JC, editor.
Voice therapy: Clinical studies. San Diego, CA: Singular;
2000. p. 46
61.
9. Stemple JC, Lee L, DAmico B, Pickup B. Efficacy of vocal
function exercises as a method of improving voice production.
J Voice. 1994;8:271
8.
10. Titze IR. Voice training and therapy with a semi-occluded
vocal tract: Rationale and scientific underpinnings. J Speech
Lang Hear Res. 2006;49:448
59.
11. Seifert U. Logopadische Hinweise zum richtigen Gebrauch
der Befehls- und Kommandosprache. Institut fur Sonderschulwesen der Humboldt-Universitat zu Berlin, Abteilung Sprachheilpadagogik; 1964.
12. Weiss DA. Introduction to functional voice therapy. Basel: S.
Karger; 1971.
13. Sovijarvi A. Die Bestimmung der Stimmkategorien mittels
Resonanzrohren. Verh. 5. int Kongr Phon Wiss, Munster
1964. Basel/New York: S. Karger; 1965.
14. Gundermann H. Die Behandlung der gestorten Sprechstimme. Fischer: Stuttgart; 1977.
aniterapiatekniikoita [Tehcniques for voice
15. Rauhala R. A
therapy]. Jyvaskyla: Finnish Reading Association; 1982. (In
Finnish)
16. Tapani M. Resonaattoriputki toiminnallisen aanihairion hoitomenetelmana. Seitseman naispotilaan seurantatutkimus
[Resonance tube as a therapy method for a functional
disorder. A follow-up study of seven female patients] [MA
thesis]. Helsinki (Finland); University of Helsinki, Department of Phonetics: Logopedics; 1992. (In Finnish)
17. Simberg S. The resonance tube*a versatile device in voice
therapy. In: Kjaer BE, editor. Nine papers on Logopedics and
Phoniatrics from 5th Nordic Congress of Logopedics and
Phoniatrics, Helsinki 2000. Odense: Forlaget Audiologopdisk Forening. p. 81
5.
18. Simberg S. Prevalence of Vocal Symptoms and Voice Disorders among Teacher Students and Teachers and a Model of
Early Intervention [Academic dissertation]. Helsinki (Finland); University of Helsinki, Department of Speech
Sciences; 2004.
19. Laukkanen A-M. Voiced bilabial fricative /B:/ as a vocal
exercise. Scand J Logoped Phoniatr. 1992;17:181
9.
20. Sovijarvi A. Nya metoder vid behandlingen av rostrubbningar. Tale og Stemme. 1969;3:121
31.
21. Bele IV. Artificially lengthened and constricted vocal tract in
vocal training methods. Logoped Phoniatr Vocol. 2005;30:
34
40.
/

156

I. R. Titze & A.-M. Laukkanen

22. Spei G. Kurze Anleitung zur Erlernung einer richtigen Ton

bildung in Sprache und Gesang. Frankfurt am Main: Aufl.
Alt; 1904.
23. Stein L. Sprach- und Stimmstorungen und ihre Behandlung
in de taglichen Praxis. Vienna-Leipzig-Bern: Weidmann &
Co; 1937.
24. Bickley CA, Stevens KN. Effects of a vocal tract constriction
on the glottal source: Data from voiced consonants. In: Baer
T, Sasaki C, Harris K, editors. Laryngeal function in
phonation and respiration. Boston, Toronto and San Diego:
College-Hill Press; 1987. p. 239
54.
25. 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 Logoped
Phoniatr. 1992;17:151
61.
26. Laukkanen A-M, Lindholm P, Vilkman E. Phonation into a
tube as a voice training method. Acoustic and physiologic
observations. Folia Phoniatr Logop. 1995;47:331
8.
27. Laukkanen A-M, Lindholm P, Vilkman E, Alku P, Haataja K.
A Physiological and acoustical study on voiced bilabial
fricative /B:/ as a vocal exercise. J Voice. 1996;10:67
77.
28. Laukkanen A-M, Lindholm P, Vilkman E. On the effects of
various vocal training methods on glottal resistance and
efficiency. A preliminary report. Folia Phoniatr Logop.
1995;47:324
30.
29. Laukkanen A-M, Lindholm P, Vilkman E. Vocal exercising
and speaking related changes in glottal resistance. A pilot
study. Logoped Phoniatr Vocol. 1998;23:85
92.
30. Story BH, Laukkanen A-M, Titze IR. Acoustic impedance of
an artificially lengthened and constricted vocal tract. J Voice.
2000;14:455
69.
31. Rothenberg M. Source-tract acoustic interaction and voice
quality. In: Lawrence VL, editor. Transcripts of the 12th
Symposium Care of Professional Voice, Part I. New York:
The Voice Foundation; 1983. p. 25
31.
32. Fant G, Lin Q. Glottal source-vocal tract acoustic interaction.
Speech Transmission Laboratory, Quarterly Progress and
Status Report, Royal Institute of Technology, Stockholm,
Sweden. 1987;1:13
27.
33. Titze IR. Theoretical analysis of maximum flow declination
rate versus maximum area declination rate in phonation. J
Speech Lang Hear Res. 2006;49:439
47.
34. Titze IR. The physics of small-amplitude oscillation of the
vocal folds. J Acoust Soc Am. 1988;83:1536
52.
35. Rothenberg M. Acoustic reinforcement of vocal fold vibratory
behavior in singing. In: Fujimura O, editor. Vocal fold
/

37.

39.

40.

41.

42.

43.
44.
45.

46.

47.

48.

49.

38.

36.

physiology, volume 2. Vocal physiology, Voice production,

Mechanisms and Functions. New York: Raven Press; 1988. p.
379
89.
Titze IR, Story BH. Acoustic interactions of the voice source
with the lower vocal tract. J Acoust Soc Am. 1997;101:2234

43.
Titze IR, Finnegan EM, Laukkanen A-M, Jaiswal S. Raising
Lung Pressure and Pitch in Vocal Warm-ups: The Use of
Flow-Resistant Straws. Journal of Singing. 2002;58:329
38.
Story BH, Titze IR. Voice simulation with a body-cover
model of the vocal folds. J Acoust Soc Am. 1995;97:1249
60.
Titze IR, Story BH. Rules for controlling low-dimensional
vocal fold models with muscle activities. J Acoust Soc Am.
2002;112:1064
76.
Liljencrants J. Speech synthesis with a reflection-type analog
[Doctoral dissertation]. Stockholm: Royal Institute of Technology, Department of Speech Communication and Music
Acoustics; 1985.
Story BH. Speech simulation with an enhanced wave-reflection model of the vocal tract [PhD dissertation]. University of
Iowa; 1995.
Story BH, Titze IR, Hoffman EA. Vocal tract area function
from magnetic resonance imaging. J Acoust Soc Am. 1996;
100:537
54.
Flanagan J. Speech analysis, synthesis and perception. 2nd
ed., expanded. Berlin: Springer Verlag; 1972.
Schutte HK. Efficiency of voice production. Groningen:
Kemper; 1980.
Story BH, Hoffman EA, Titze IR. Vocal tract imaging: A
comparison of MRI and EBCT. In: Hoffman EA, Editor.
Medical Imaging 1996: Physiology and Function from Multidimensional Images. Proceedings of SPIE Volume 2709.
Newport Beach, CA: SPIE; 1996. p. 209
22.
Titze IR. Acoustic interpretation of resonant voice. J Voice.
2001;15:519
28.
Titze IR. Regulating glottal airflow in phonation: application
of the maximum power transfer theorem to a low dimensional
phonation model. J Acoust Soc Am. 2002;111:367
76.
Titze IR. Theory of glottal airflow and source-filter interaction in speaking and singing. Acta Acustica-Acustica. 2004;
90:641
8.
Berry D, Verdolini K, Montequin DW, Hess MM, Chan RW,
Titze IR. A quantitative output-cost ratio in voice production.
J Speech Lang Hear Res. 2001;44:29
37.
Appelman DR. The Science of Vocal Pedagogy: Theory and
Application. Bloomington, IN: Indiana University Press;
1967.
/

50.