Академический Документы
Профессиональный Документы
Культура Документы
*Eliana Midori Hanayama, †Zuleica Antonia Camargo, ‡Domingos Hiroshi Tsuji, and §Sı́lvia Maria Rebelo Pinho,
São Paulo, Brazil
Summary. The metallic voice is usually confused with ring or nasality by singers and nontrained listeners, who are
not used to perceptual vocal analysis. They believe a metallic voice results from a rise in fundamental frequency. A di-
agnostic error in this aspect may lead to lowering pitch, an incorrect procedure that could cause vocal overload and fa-
tigue. The purpose of this article is to study the quality of metallic voice considering the correlation between information
of the physiological and acoustic plans, based on a perceptive consensual assumption. Fiberscopic video pharyngolar-
yngoscopy was performed on 21 professional singers while speaking vowel [e]—in normal and metallic modes to ob-
serve muscular movements and structural changes of the velopharynx, pharynx, and larynx. Vocal samples captured
simultaneously to the fiberscopic examination were acoustically analyzed. Frequency and amplitude of the first four
formants (F1, F2, F3, and F4) were extracted by means of linear predictor coefficients (LPC) spectrum and were statis-
tically analyzed. Vocal tract adjustments such as velar lowering, pharyngeal wall narrowing, laryngeal rise, aryepiglot-
tic, and lateral laryngeal constrictions were frequently found; there were no significant changes in frequency and
amplitude of F1 in the metallic voice; there were significant increases in amplitudes of F2, F3, and F4 and in frequency
for F2; metallic voice perceived as louder was correlated to an increase in amplitude of F3 and F4. Physiological adjust-
ments of velopharynx, pharynx, and larynx are combined in characterizing the metallic voice and can be acoustically
related to changes in formant pattern.
Key Words: Voice quality–Voice disorders–Speech acoustics–Speech perception.
INTRODUCTION One of the polemical issues that involve vocal quality is the
The scarce data in the literature about vocal performance metallic voice, usually confused with ring, nasality, or high
patterns and their correlation with physiological and acoustic pitch by singers and lay people, who are not used to perceptual
information make it difficult to establish efficient tools for vocal vocal analysis.7 The occurrence of voice metallization was as-
description, assessment, and training. sociated to cul-de-sac resonance in cases of hearing impair-
Vocal quality refers to the perceptive aspect of the voice in ment. There is proximity between these attributes, which
the psychological sphere, comprising several attributes that makes the occurrence of mistaken diagnostic assessments
are presented in a continuum and are described with the use very common.8
of terms based on auditive perception in each one of them, Barnes, in his thesis, analyzed the vocal quality of students
such as normal/abnormal, good/bad, bright/dark, oral/nasal, from the University of Iowa in 1932, and found 27.2% of
from one extreme to the other.1 The physical bases of the differ- them had a metallic voice.9
ences between the several vocal qualities correspond to a set of The impressionistic term metallic voice is a label that has
adjustments in the vocal tract.1,2 been used to characterize a strident, irritating, penetrating,
From the phonetic point of view, vocal quality is determined weeping, and fine voice,9 a rough voice,10 and as bright, clean,
by the present acoustic parameters, related to the vibration of sharp, biting voice.2
the vocal folds and configuration of the supraglottic vocal Metallization was related to hypertonicity of pharyngeal con-
tract.2 strictors and to the elevation of the larynx, which decrease both
Aerodynamic, acoustical, and perceptual studies compared the length and the width of the pharynx11,12 and contract the
different vocal qualities in singing such as belt, twang, opera,3 walls of the vocal tract. The walls, in turn, become a rigid sur-
country western, operatic, and Broadway vocal styles,4,5 pop, face that emphasizes high acoustic frequencies, thus generating
jazz, and blues6 and have found distinguished acoustical fea- a tense, metallic, strident voice.13 The resonance focus becomes
tures among different vocal qualities. pharyngeal. In addition to pharyngeal constrictions, there
seems to be an elevation in the back of the tongue, positioning
Accepted for publication December 18, 2006.
of the lips in a smiling posture, and aryepiglottic constric-
Presented at XII Brazilian Congress of Speech Therapy, October 7, 2004, Iguassu Falls, tion,7,14 and velar lowering.12
ISSN 1807-3115.
From the *Center for Specialization in Clinical Speech Pathology and Audiology, CE-
Speakers may consider the metallic voice an undesirable
FAC, São Paulo, Brazil; yIntegrated Laboratory on Acoustic Analysis and Cognition, characteristic, because it is considered sharp and irritating.2
LIAAC of the Department of Linguistics of the School of Communication and Philosophy
of the Pontifical Catholic University of São Paulo, São Paulo, Brazil; zDivision of Otorhi-
However, it may be an efficient resource for vocal projection
nolaryngology Clinic, Clinics Hospital, School of Medicine, University of São Paulo, São in the theater and in noisy environments with poor acoustics,
Paulo, Brazil; and the xINVOZ, Voice Institute: Communication and Professional Voice,
São Paulo, Brazil.
giving the actor confidence,12 in addition to being useful as
Address correspondence and reprint requests to Eliana Midori Hanayama, R Mario Para- a dramatic resource in the characterization of a character, in
nhos Pederneiras, 156, São Paulo, SP, Brazil, CEP 04072-060. E-mail: midorihanayama@
gmail.com
some types of singing, such as the American and the Brazilian
Journal of Voice, Vol. 23, No. 1, pp. 62-70 country music.7
0892-1997/$36.00
Ó 2009 The Voice Foundation
The present study examined the metallic vocal quality from
doi:10.1016/j.jvoice.2006.12.006 the standpoint of the correlations between the information of
Eliana Midori Hanayama, et al Metallic Voice 63
the physiological and acoustic plans, based on a perceptive con- was obtained. The laryngeal sequence added a laryngostrobo-
sensual assumption. scopy to record the movement of the vocal folds in detail.
using the CSL software for Windows of the Voice and Speech different relative postures of the structures of the vocal tract
Laboratory of the Center for Specialization in Clinical Speech were qualitatively analyzed.
Pathology and Audiology (CEFAC), in the sampling frequency The analysis of the physiological data involved the detection
in 22 050 Hz, 16 bits. For sample analysis, the procedures per- or nondetection of the adjustments of the structures of the vocal
formed included acoustic inspection (wave format and wide- tract15 during oral and metallic mode utterances, based on the
band spectrogram), in which extracts of utterance stability comparison of the traces of photographic records, regarding
were selected to obtain the measures of frequency (Hz) and am- the soft palate, presence or absence of slight tension, presence
plitude (dB) of the four first formants (F1, F2, F3, and F4). The or absence of moderate tension, presence or absence of lower-
wideband spectrograms were generated with Hanning win- ing, medialization of the pharynx walls, changes in the height of
dows, defining the extension of the superior frequency limit the larynx, aryepiglottic tightening, and lateral larynx tighten-
at 5 kHz because higher frequency energy can lead to inaccu- ing. When a certain adjustment was detected, a ‘‘1’’ was
rate allocation of zeroes in the error prediction filter. The dura- marked, when not, a ‘‘0.’’
tion of the window was defined at 10 milliseconds. When the anatomical structures were not visible, the image
Downsampling to 11 025 Hz was made and the filter order was classified as ‘‘nonvisible.’’
was kept between 10 and 12 to obtain the spectral envelope.16
The intensity values of spectral peaks were normalized to Statistical analysis
proportional values to eliminate variations due to different For the analysis of frequency acoustic data and formant ampli-
levels of recording and different loudnesses of utterance at col- tude data (F1, F2, F3, and F4) of the oral mode and metallic
lection.17,18 In this manner, the values referring to F1 intensity mode, and comparison per gender, we used the Student’s t
were normalized to value 100 and, the others, to fractions of test for matching data, Mann-Whitney test, and Fisher Exact
100.19 Test. To check the correlation between the data referring to
the formants and the adjustments of the vocal tract, we used
Spearman correlation analysis. The significance level consid-
ered was 5%.
Physiological analysis
For each sequence of nasofibroscopic images, both for the oral
mode utterance and metallic mode utterance, we selected and RESULTS
froze the images that presented the highest degree of constric- Table 2 shows the analysis of the nasopharyngolaryngofiber-
tion of all vocal tract structures involved in that segment, for scopic images observed during metallic mode utterance on
the three positions recorded: velopharynx, pharyngolarynx, three analysis positions: velopharyngeal, laryngopharyngeal,
and larynx. All selected images were printed on film by and laryngeal sequences, in each subject.
a Sony video image printer, model UP2300 (Japan), and were The frequencies of adjustments (in number of occurrences
then traced in transparencies, for the comparative analysis of and %) seen are presented in Table 3. Most showed aryepiglot-
the movement of the structures. tic constriction (81%) and laryngeal raising (71%). Among ve-
For this analysis, two situations were defined: the initial po- lopharyngeal adjustments, the most frequent was lowering
sition, observed during the oral mode utterance, and the analy- (48%). Other relatively frequent adjustments included lateral
sis position, observed during the metallic mode utterance.15 The constriction (38%) and light tightening (24%). In nine subjects
TABLE 2.
Analysis of Nasopharyngolaryngofiberscopic Images Observed During Metallic Mode Utterance: Velopharyngeal,
Pharyngeal, and Laryngeal Movements in Each Subject
Sequence 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Velopharynx
No movement 1 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 1 0 0
Light tightening 0 1 0 0 0 0 0 0 1 0 1 0 0 0 1 1 0 0 0 0 0
Moderate tightening 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0
Lowering 0 0 0 1 1 1 1 1 0 1 0 1 1 0 0 0 0 0 0 1 1
Laryngopharynx
Pharynx: medialization NV NV NV 1 1 0 1 NV 0 1 NV NV 1 1 NV NV 1 NV 1 1 1
Larynx: raising 1 1 1 1 1 0 1 1 0 1 0 1 1 0 0 0 1 1 1 1 1
Larynx: lowering 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0
Larynx
Aryepiglottic tightening 1 1 1 1 1 1 1 1 0 1 0 1 1 1 1 1 0 1 1 1 1
Lateral tightening 0 0 0 0 0 0 1 0 0 0 1 1 1 1 1 0 1 0 0 1 1
0, no adjustment in analysis position; 1, adjustment present in analysis position; NV, not visible.
Eliana Midori Hanayama, et al Metallic Voice 65
TABLE 3.
Adjustments Observed in the Vocal Tract (in Number of
Occurrences and %) During Metallic Mode Utterance of
the Subjects
Vocal Tract Adjustment n %
Velopharynx
No movement 3 14
Light tightening 5 24
Moderate tightening 3 14
Lowering 10 48
Laryngopharynx
Pharynx: medialization 10 48
Larynx: raising 15 71
Larynx: lowering 2 9
Larynx
Aryepiglottic tightening 17 81
Lateral tightening 8 38
metallic mode voice defined as having pharyngeal resonance to obtain images of the many positions during the same utter-
and strident quality, according to the descriptions of Boone ance.
and McFarlane,11 Berry,12 and Pinho.7 The supraglottic adjustments observed during metallic mode
Vocal tract adjustments, determining for the vocal quality of utterance presented a relevant occurrence in this study. Aryepi-
utterance, have been studied in the phonetic model, as a combi- glottic constriction was seen in 81% of the subjects. This adjust-
nation of intrinsic factors, related to the anatomy unique to a cer- ment was observed during singing in the twang, belting, and
tain speaker, and extrinsic, related to long-term muscular opera styles, where the presence of the singer’s formant was
adjustments. These adjustments may be present in normal sys- verified.3 The main function of the aryepiglottic fold was
tems,2 as a vocal resource, or as a compensation to changes. thought to be related to the epiglottis in swallowing but it was
Some supraglottic adjustments were analyzed in this study also related to speaking, when those folds become high,
through the direct observation of the involved structures and smooth, and taut, producing a deep supraglottic cavity for the
acoustic analysis. regulation of resonance.13
Most subjects in this study, professional singers, showed that Lateral constriction was observed in 38%. The ventricle of
they knew the vocal quality under study and produced it follow- Morgagni provides a space for free vibration of the vocal
ing verbal command. fold, and it seems that it functions as a resonator. When pressed,
Simultaneous images of the velopharynx, pharynx, and lar- its volume was reduced by the ventricular folds, and this was re-
ynx could not be obtained only through the nasopharyngolar- lated to the lack of suppression of sharp harmonics, which
yngoscopy. Adjustments made it difficult for the larynx to be would result in a rough and metallic voice. This phenomenon
simultaneously observed. In this manner, it was not possible would be related to vocal potency in singers who have small
Eliana Midori Hanayama, et al Metallic Voice 67
TABLE 4.
Average Normalized Values of Levels of Formants (F1, F2,
F3, and F4) + SD in dB During Oral and Metallic Mode
Utterance of Subjects
Amplitude (dB) Oral Metallic P
F1 14 ± 5 16 ± 5 0.126
F2 71 ± 10 85 ± 11 <0.001
F3 63 ± 11 80 ± 10 <0.001
FIGURE 6. Comparison of normalized amplitude variation of F3 in
F4 46 ± 12 63 ± 17 <0.001
dB between oral and metallic mode emissions for each subject.
68 Journal of Voice, Vol. 23, No. 1, 2009
TABLE 5.
Average Frequency Values of Formants (F1, F2, F3, and
F4) + SD in Hz During Oral and Metallic Mode Emission of
Subjects
Frequency (Hz) Oral Metal P
F1 443 + 84 443 + 68 >0.999
F2 2038 ± 245 2186 ± 250 <0.001
F3 2708 ± 217 2840 ± 341 0.052
F4 3872 ± 425 3973 ± 510 0.158
FIGURE 9. LPC spectrum of oral and metallic emission—subject 3.
Eliana Midori Hanayama, et al Metallic Voice 69
It is worth emphasizing that velar lowering does not mean ve- would be perfect and would greatly help the understanding of
lopharyngeal closing deficit, which can be confirmed by the fact the behavior of F2, leaving, therefore, a gap to be explained
that no attenuation of formants in the acoustic analysis was ob- by future research.
served. In fact, velopharyngeal opening that could be caused by Lindblom and Sundberg37 reported effects of frequency low-
velar lowering is not also necessarily related to nasal quality of ering of all formants to the low larynx position, with a more pro-
the vowel timbre.30 nounced effect in F2. In theory, the opposite situation, that is,
There are almost always other phenomena parallel to pharyn- larynx raising, could be related to an increase in F2. However,
geal constriction, such as change in laryngeal height and ten- the laryngeal raising observed in the present study did not cor-
dency to velar lowering, which may cause a certain nasalizing relate significantly to increase in F2.
feeling.2 This should be related to the frequent confusion in No significant differences were detected in the data observed
the perceptive evaluation of the metallic quality with nasaliz- when comparing male and female subjects. The only variable
ing. The metallic quality, sometimes mistaken for nasalizing, with a significant difference in a statistical test, although not
may be perceived in western country singing and in Brazilian confirmed by another, was larynx lowering, observed in two
country singing.3 of seven male subjects, and it was not observed in any female
The vocal quality is clearly influenced by the rising ampli- subject. Future research should assess the differences between
tude of F2 and F3, when compared to the wider amplitude of men and women in the incidence of this kind of adjustment.
F1.31 The metallic voice is commonly associated with the raise in
The voices of classical singers and the speech of many indi- regular fundamental frequency by nontrained listeners. A diag-
viduals with ring are characterized by an increased amplitude in nostic error in this aspect leads to an incorrect therapeutic pro-
F3 and F4.19,32 The fact that the sample of the present study had cedure of lowering regular frequency, which in turn would
a significant rise in amplitude of both formants may indicate cause vocal overload and fatigue.7 The factors that in fact deter-
that the ring component accompanies the metallic voice. mine the high pitch perceived by the listener, and that could be
Another phenomenon related to ring is the potency observed observed in this study, could be related to the emphasis in the
in lyrical singing. Professional classical singers have certain re- amplitude in the high spectral region of F2, F3, and F4.
inforced frequency bands, around 3 kHz, the singer’s formant, The functional diagnosis should, therefore, consider that the
enabling him/her to be heard, perfectly, even while accompa- physiological correspondences to the perceptive-auditory find-
nied by a symphonic orchestra, and it is acquired through tech- ing of the metallic mode were pharyngeal constriction and func-
nique and experience.33–35 Laboratory experience has related tional reduction of vocal tract cavities due to tension of
this phenomenon to an anteroposterior glottis narrowing also structures14 and also of remaining adjustments observed in
observed in the present study.21 this study. In the future, lingual and labial adjustments, not an-
In the country style, on the other hand, in which the voice is alyzed in this study, should be taken into consideration.
used with a feature similar to spoken voice, singers do not have The use of metallic voice in some singing situations or styles
the singer’s formant36 despite the tendency to raise the fre- and the adjustments observed in this study, especially epilarynx
quency of F2. Voices considered efficient for professional use narrowing, seem to be the most effective configurations to be
were related, in acoustic analyses, to the concentration between used, from the vocal hygiene point of view.
3 and 4 kHz frequencies, called ‘‘speaker’s formant.’’17
As to the behavior of the frequencies of formants in the me-
tallic mode, there was a significant shift of F2 frequency to
a higher region. Formants F1, F3, and F4 did not change signif- CONCLUSION
icantly in frequency. F2 is extremely sensitive to anteroposterior Regarding physiological and acoustic characteristics of the me-
positioning of the tongue. The tongue anteriorizing movement tallic mode in relation to the oral mode:
that narrows the region of the hard palate and expands the pha-
ryngeal cavity leads to an elevation of this formant.23,37 When 1. There was a relevant incidence of adjustments of the vo-
an adult woman adopts a ‘‘little girl’s’’ voice, the elevation is cal tract such as velar lowering, pharyngeal wall medial-
mentioned and may be a slight tongue anteriorizing and, some- ization, laryngeal raising, aryepiglottic, and lateral
times, laryngeal raising.2 In this situation, the frequency of F2 tightening.
becomes very high and near F3, a behavior that was observed 2. There was no significant change in F1, neither in relation
in the present study. Probably the significant frequency rise of to frequency nor in relation to amplitude.
F2 observed in the metallic mode may be related to this kind 3. There was a significant change in the amplitude of F2, F3,
of tongue adjustment. The adjustment was not evaluated in and F4.
this study, although it may have occurred. Tongue dislocation 4. There was a significant dislocation of F2 in the direction
is related to extremely complex synergetic muscular move- of high frequencies.
ments. Findings of tests such as MRI or 3D images captured 5. Higher intensity was significantly correlated with the el-
by electron beam computed tomography, where the vocal tract evation in the amplitude of F3 and F4.
configuration is a function of changes in vocal modes, can be
assessed by comparing differences in area functions along the The analyses performed in this study provided data that may
length of the vocal tract29 allied to filming lip behavior. This help toward the correct approach of this vocal mode and will
70 Journal of Voice, Vol. 23, No. 1, 2009
support future studies that should be included in the analysis of 18. Nordenberg M, Sundberg J. Effect on LTAS of vocal loudness variation.
the glottis source and of labial and lingual adjustments. Logop Phoniatr Vocol. 2004;29:183-191.
19. Oliveira-Barrichelo VM, Heuer RJ, Dean CM, Sataloff RT. Comparison of
singer’s formant, speaker’s ring, and LTA spectrum among classical singers
REFERENCES and untrained normal speakers. J Voice. 2001;15:344-350.
1. Colton RH, Estill JA. Elements of voice quality: perceptual, acoustic and 20. Ekholm E, Papagiannis GC, Chagnos FP. Relating objective measurements
physiologic aspects. Speech and Language: Advances in Basic Research to expert evaluation of voice quality in western classical singing: critical
and Practice. New York, NY: Academic Press Inc.; 1981. 311–403. perceptual parameters. J Voice. 1998;12:182-196.
2. Laver J. The Phonetic Description of Voice Quality. Cambridge, MA: Cam- 21. Titze IR, Story BH. Acoustic interactions of the voice source with the lower
bridge University Press; 1980. vocal tract. J Acoust Soc Am. 1997;101:2234-2243.
3. Yanagisawa E, Estill J, Kmucha ST, Leder SB. The contribution of aryepi- 22. Bergan CC, Titze IR, Story B. The perception of two vocal qualities in
glottic constriction to ‘‘Ringing’’ voice quality—a videolaryngoscopic a synthesized vocal utterance: ring and pressed voice. J Voice. 2004;18:
study with acoustic analysis. J Voice. 1989;3:342-350. 305-317.
4. Stone RE, Cleveland TF, Sundberg J, Prokop J. Aerodynamic and acousti- 23. Fant G. Acoustic Theory of Speech Production. Paris: Mouton; 1970.
cal measures of speech, operatic, and Broadway vocal styles in a profes- 24. Henrich N, D’Alessandro C, Castellengo M, Doval B. Glottal open quotient
sional female singer. Speech, Music and Hearing. In: TMH-QPSR, Vol. in singing: measurements and correlation with laryngeal mechanisms, vo-
43. Stockholm, Sweden: KTH; 2002. 17–29. cal intensity, and fundamental frequency. J Acoust Soc Am. 2005;117:
5. Stone RE, Cleveland TF, Sundberg J. Acoustic and aerodynamic character- 1417-1430.
istics of country-western, operatic and Broadway singing styles compared 25. Baken RJ. Electroglottography. J Voice. 1992;6:98-110.
to speech. J Acoust Soc Am. 2003;113:2242-2243. 26. Greene MCL. The Voice and Its Disorders. Philadelphia, PA: Pitman Med-
6. Sundberg J, Thalén M, Alku P, Vilkman E. Estimating perceived phonatory ical; 1972.
pressedness in singing from flow glottograms. Speech, Music and Hearing. 27. Sundberg J. Research on the singing voice in retrospect. Speech, Music and
In: TMH-QPSR, Vol. 43. Stockholm, Sweden: KTH; 2002. 89–96. Hearing. In: TMH-QPSR, Vol. 45. Stockholm, Sweden: KTH; 2003. 11–22.
7. Pinho SMR. Avaliação e Tratamento da Voz. Fundamentos em Fonoaudio- 28. Hardcastle WJ. Physiology of Speech Production. London: Academic
logia. Rio de Janeiro: Guanabara Koogan; 1998. 3–37. Press; 1976.
8. Wilson DK. Children’s voice problems. Voice Problems of Children. 3rd ed. 29. Tom K, Titze IR, Hoffman EA, Story BH. Three-dimensional vocal tract
Maryland: Williams & Wilkins; 1987. 2–15. imaging and formant structure: varying vocal register, pitch, and loudness.
9. Van Dusen CR. A laboratory study of the metallic voice. J Speech Dis. J Acoust Soc Am. 2001;109:742-747.
1941;6:137-140. 30. Birch P, Gumoes B, Stavad H, Prytz S, Bjorkner E, Sundberg J. Velum be-
10. Van den Berg JW. On the role of the laryngeal ventricle in voice production. havior in professional classic operatic singing. J Voice. 2002;16:61-71.
Folia Phoniatr. 1955;7:57-69. 31. Borden GJ, Harris KS, Raphael LJ. Speech Science Primer—Physiology,
11. Boone DR, McFarlane SC. The Voice and Voice Therapy. 4th ed. New Jer- Acoustics and Perception of Speech. 3rd ed. Baltimore, MD: Williams &
sey: Prentice-Hall; 1988. Wilkins; 1994.
12. Berry C. Voice and the Actor. New York, NY: MacMillan; 1991. 32. Sundberg J. Vocal tract resonance. In: Sataloff RT, ed. Professional Voice:
13. Kaplan HM. Anatomy and Physiology of Speech. New York, NY: The Science and Art of Clinical Care. New York, NY: Raven Press Ltd.;
McGraw-Hill; 1971. 1991:49-68.
14. Pinho SMR. Fonoterapia nos distúrbios vocais. In: Campos CAH, 33. Sundberg J. Articulatory interpretation of the ‘‘singing formant.’’ J Acoust
Costa HOO, eds. Tratado de Otorrinolaringologia. São Paulo: Roca; Soc Am. 1974;55:838-844.
2002:596-605. 34. Sundberg J. Level and center frequency of the singer’s formant. J Voice.
15. Lovetri J, Lesh S, Woo P. Preliminary study on the ability of trained singers 2001;15:176-186.
to control the intrinsic and extrinsic laryngeal musculature. J Voice. 35. Pillot C, Quattrocchi S. Acoustic measurements, perceptive judgements
1999;13:219-226. and physiological correlates of the ‘‘singing-formant’’ in male and female
16. Vieira MN. Automated measures of dysphonias and the phonatory effects of opera singers. Rev Laryngol Otol Rhinol (Bord). 1996;117:335-339.
asymmetries in the posterior larynx [doctorate]. Edinburgh: University of 36. Cleveland TF, Sundberg J, Stone RE. Long-term average characteristics of
Edinburgh; 1997. country singers during speaking and singing. J Voice. 2001;15:54-60.
17. Nawka T, Anders LC, Cebulla M, Zurakowski D. The speaker’s formant in 37. Lindblom EF, Sundberg JEF. Acoustical consequences of lip, tongue, jaw,
males voices. J Voice. 1997;11:422-428. and larynx movement. J Acoust Soc Am. 1971;50:1166-1179.