Sie sind auf Seite 1von 12

Journal of Voice

Vol. 15, No. 4, pp. 457-468


2001 The Voice Foundation

Singers Formant in Sopranos: Fact or Fiction?


*Rudolf Weiss, W.S. Brown, Jr., and Jack Morris
*Linguistics Program, Western Washington University, Bellingham, Washington; Department of
Communication Sciences and Disorders, University of Florida, Gainesville, Florida; Music
Department, Western Washington University, Bellingham, Washington

Summary: Although it is generally agreed that the singers formant (FS) is a


prerequisite for successful stage performance, the results of this research do not
support its presence in the soprano voices of trained female singers. Results are
based on a recent investigation testing 10 advanced/professional sopranos in
two groups singing sustained vowels at three frequencies: high (932 Hz), mid
(622 Hz), and low (261 Hz). Spectrographic analysis shows that the nature of
harmonic energy varies in relation to pitch. A resonance band somewhat resembling the tenor FS was usually evident in vowels sung at low and mid pitch.
However, unlike the FS of typically less than 1 kHz bandwidth associated with
tenors, sopranos singing similar pitches produced corresponding bandwidths
which were significantly broader, usually at least 2-kHz wide. Vowels sung by
sopranos at high-pitch levels exhibited strong fundamental frequency production with strong reinforcement of adjacent harmonics extending to 5 kHz and
beyond. This type of production in essence nullifies the necessity for a typical
FS. Absence of the FS in strong soprano voices might also imply the adaptation
of a sufficiently different overall vocal tract configuration, so that techniques
geared to developing maximal projection should not be the same as those developed to maximize the FS in other voices. Key Words: FormantsSingers
formantVocal resonanceSopranosSinging.

enlarging the vocal tract through whatever means


available. Yet, it has not been proved that: (1) the FS
exists or can exist in all classically trained voices; (2)
we definitively know what its physiological manifestation is when it does exist; (3) it is produced in a
similar fashion where/when it exists; and (4) it is uniformly demonstrable in the acoustic domain. It could
also be asserted that male and some lower female
voices tend to have and need an FS, whereas high voices may or may not have it nor need it.
The FS has been the subject of numerous discussions and debates over the course of many years.
Questions as to its necessity, requirement, frequency,
and variability have been raised in professional voice
journals and meetings for more than half a century.

INTRODUCTION
The singers formant (FS) remains an unresolved
issue in voice circles. Very few contemporary trainers
of advanced voices would argue that the FS does not
exist. In fact, techniques directed at facilitating its development have been and continue to be widely used.
These include attempts at lowering the larynx and
Accepted for publication April 30, 2001.
Presented at the 28th Symposium: Care of the Professional
Voice, Philadelphia, Pennsylvania, June 26, 1999.
Address correspondence and reprint requests to Rudolf
Weiss, PhD, Director, Linguistics Program, Hu 237, Western
Washington University, Bellingham, WA 98225-9057.
e-mail: weiss@cc.wwu.edu

457

458

RUDOLF WEISS ET AL

Schutte and Miller1 claim that Bartholomew 2 may


have been the first who singled out, via spectrographic means, that the frequency around 3 kHz is
highly strengthened or amplified. Subsequent studies
have involved acoustic analysis, electromyography,
computer enhanced parametric analysis, etc., to find
the holy grail of the singers formant. It was originally referred to only as an area of spectral reinforcement and only recently has the term singing formant 3 or singers formant 4 been used. In our past
research, we have chosen to refer to it simply as FS in
analogy to the notation for vowel formants and fundamental frequency.5,6
The following are points of general agreement regarding this phenomenon:
1. The FS exists and is reflected as strong spectral
reinforcement at around 3 kHz;57
2. The FS is primarily identified with trained male
voices;46
3. The FS is usually acquired only as a matter of
training;5,6,8
4. The FS is required to project on the modern stage;9
5. The FS is most likely also associated with the
projection of requirements for stage speakers.10,11 If so, it could also rightly be called a
speakers formant;
6. As Sundberg12 has pointed out at the 12th International Congress of Phonetic Sciences
(ICPhSc) in Stockholm, the spectral location of
the FS is both favored by the hearing curve and
fills a special hole in the typical orchestra;
7. The existence of the FS can be demonstrated
through acoustic analysis and through physiological studies;5,6
8. The precise location of the FS varies. It may be
affected by individual voice types and ranges,
the vowels attempted, the pitch and amplitude
produced, and individual physiology;6
9. Most agree that the FS is glottically related and
that it may be a resonance somehow involving
the glottic/laryngeal area.
Although much is known, there are numerous
questions about the FS where there is either insufficient evidence in the literature, or lack of consensus,
or the results are inconclusive, nonexistent, or contradictory. These questions include:
1. Does the FS exist in high womens voices? If
not, what allows them to project even more
than male voices?
Journal of Voice, Vol. 15, No. 4, 2001

2. What is the bandwidth of the FS? Does it


change? Is it an individual matter? (Schutte and
Miller1 show that the bandwidth increases with
the F0 as does its spectral location.)
3. What are the physiological ramifications favoring production of the FS? Where precisely is
the cavity or resonance area causing this spectral reinforcement?
4. Since lowering the larynx is often involved in
the attainment of FS production of stage tenors
and other male voices, is it unreasonable to assume that the supraglottic and/or subglottic areas have something to do with the production
of the FS?
5. How stable is the FS of a singer across F0
frequencies?
6. How does the strength and presence of the FS
relate to the overall quality of the voice?
Although these and other questions are worthy of
further study, our primary focus will be directed to
the question of the existence of a soprano FS. Certainly, female singers, especially strong sopranos,
have tremendous potential for projection.9 However,
contrary to some studies,8,10,13 which lay claim to
the existence of an FS in high womens voices, we
have found little evidence supporting the existence
of an FS in sopranos, at least any resembling that of
the male voices even at overlapping F0 frequencies.5,6 Further, we could find no evidence of one
when very high fundamental frequencies were involved. This is in agreement with some of the literature in this arena, particularly that of Sundberg.7
From a voice scientists point of view, the FS is an
acoustic term referring to an envelope of peak resonance in much the same way that the terms F1 and F2
are usually referred to in association with the pharyngeal and oral cavity resonance for vowel sounds.
Thus, this term must reflect the reality of acoustic
fact. That is, the FS must be evident on a spectral display as a demonstrable peak which can only be the
result of the reinforcement or amplification of certain
harmonic overtones which fall into the bandwidth of
the reinforcement potential of a singers formant
(typically 2.8 to 3.2 kHz). This is precisely the type
of evidence lacking for soprano voices. The analyst
looking for the FS finds that it soon becomes a phantom the more the range of the fundamental frequency

SINGERS FORMANT IN SOPRANOS


(F0) is extended upward. This reflects the practical solution found by the singer, i.e., the FS is usurped in that
most of the spectral energy produced by the voice at
high pitch levels is already high-frequency energy of
the fundamental and harmonics, and becomes increasingly higher as the pitch is extended. Thus, a cogent argument could be made that high female voices simply
do not need an FS. In fact, one could further argue that
strong soprano voices are not limited by an FS for maximal high-frequency projection.
However, this strong high-frequency production
and reinforcement comes at the expense of vowel
quality. It is well known that the potential for varying
vowel quality of high sopranos is, of necessity, diminished when high-pitched vowels are sung. There
may be no harmonic energy at the F1 or F2 resonance
frequencies required for the distinction of vowel
quality at all. Nor is it unlikely that there would be no
energy which would fall within the 2.83.2 kHz
range of the typical FS. It is a necessary manifestation
and consequence of super pitch voice production
that harmonics are not present where formants, the
result of cavity resonances, peak.7 There may be considerable effort to adjust the vocal tract to facilitate
formant tracking in attempting to produce some vowel quality distinction when singing at high-pitch levels, albeit with questionable success.14
A thorough discussion of the physical and physiological nature of the FS is beyond the scope of this paper. A number of theories have been proposed and
supported as explanations for the FS phenomenon.
These include, among others, such concepts as Sundbergs formant clustering,7 his proposed 1:6 ratio (laryngeal tube to vocal tract) required to produce a
strong FS ( or F4 ) resonance,15 subglottal resonance
and impedence matching,16,17 wide pharynx phenomenon,18 and the necessity for lowering of the larynx.19
Nevertheless, the focus on the 3-kHz frequency as
the key resonance area for all voices, by whatever
name, is and continues to be in many circles the essential ingredient and prerequisite of the stage voice.
This was recognized by Schultz-Coulon et al8 and
was demonstrated in Winckels10 valuable article in
1971. In the latter, reference is made to the spectral
frequency in the area of 3 kHz: Most informative is
a frequency in the area of 3,000 Hz. He identifies
this simply as F3. Moreover, this frequency compo-

459

nent has the function of a carrying power which is


important for singing in large opera houses. He compares the voices of D. Fischer-Diskau and Maria
Callas, using total energy versus energy in the vicinity of 3 kHz. Peak energy in the area in the male
voice is only 5 dB less than total energy, whereas the
difference for Maria Callas was about 20 dB less; i.e.,
in the male voice most of the energy was diverted to
the FS, in the female only part of the energy, a good
portion most likely diverted to even higher frequencies. Suffice it to say, as Hollien so aptly points out
in his informative article, That Golden VoiceTalent or Training?, the issue of the presence of the
singers formant and its physiological manifestation
are very complex.20
This brings us back to the central question: the soprano Fsfact or fiction? If, as indicated above, the
harmonic energy needed for potential FS frequency resonation is not present, nor the necessary harmonic energy to realize the resonating benefits of an F1 or even
an F2, why, then, is the production of the soprano voice
still the model of projection, especially at high fundamental frequencies? Are there other factors more
prominent than an FS that allow the voice to project so
effectively? Furthermore, what can be determined
about the mid and low frequencies of the soprano voice
where the spacing of harmonic overtones more closely
resembles those of the higher male voices and altos?
PURPOSE OF THE STUDY
In a previous study,6 we were generally able to
demonstrate both the presence of the FS and the degree to which its frequency is affected by intensity,
pitch, and vowel quality. We determined average FS
frequencies for basses (2.83 kHz) and for tenors
(3.29 kHz). Female voices, particularly those of the
sopranos, often revealed evidence of some energy in
partials somewhere in the 3 to 3.5 kHz region (as in
Schultz-Coulton et al8), but, most often, also found
significant reinforcement of higher harmonics as
well. Our measurements for sopranos showed peaks
(similar to those produced by the FS) from 2.6 to 4.6
kHzmuch too variable when compared to the
malesthus leading us to the conclusion that the FS
as a 3-kHz reinforcement among soprano voices is
really not a significant phenomenon. What seemed to
Journal of Voice, Vol. 15, No. 4, 2001

460

RUDOLF WEISS ET AL

be significant is simply broad reinforcement of highfrequency harmonics. This prompted us to expand


upon a segment of our former study, namely to determine, if we could, whether the FS exists in sopranos. If not, then to attempt to determine the true nature of the strong female voice, the ring.
PROCEDURE
Experimental design
Since results of our previous study are based on the
participation of only four advanced sopranos, the
current study undertakes an expansion of an additional six sopranos. Attempts were made to replicate
the conditions of the first experiment 6 and as closely
as possible three years later using the Institute for the
Advanced Study of the Communication Processes
facilities at the University of Florida. The data of the
four advanced soprano singers in the first study
would be reanalyzed and compared to the results obtained with the six sopranos in the second study. In
both studies the cardinal vowels in the order of [i e a
o u] were recorded on three singing notes:
low pitch: C (261 Hz) mid pitch: Eb (622 Hz)
high pitch Bb (932 Hz)
The conditions differed, however, in the following
respects. The first group recorded at Western Washington University (WWU) was room-miked in a studio approximately 14 ft. by 12 ft. with microphones
placed at a distance of six feet. The participants in the
second study at the University of Florida (UFL) were
close-miked in a smaller practice room of approximately 10 ft. by 10 ft. with a head-microphone placed
2 to 3 inches from the mouth. This was done in the
hopes of eliciting more accurate information about
the voice with less interference from room acoustics
than had been achieved in the first study where room
mikes had been used. Additionally, we were able to
use digital audiotape (DAT) equipment in procuring
what we hoped would be even more accurate data in
the second study. In both situations, singers were given a warm-up period prior to recording in which the
recorder was set to peak as closely as possible to the
zero level of the recording device during the loudest
production by the individual singer. Once the level
was set, it remained unchanged for the rest of the participants recording session.
Journal of Voice, Vol. 15, No. 4, 2001

Equipment
WWU study
Recordings were made at the facilities of the Music
Department at Western Washington University between
January and June 1995. A Marantz (Marantz Corporation, Aurora, IL) 1120 integrated amplifier, a Technics
(Matsushita Electric Corporation of America, Secaucus,
NJ) Model M227k tape deck, and Realistic (Radio
Shack Corporation, Fort Worth, TX) Commercial Model 33.10708 microphones were the equipment used.
UFL study
Recordings were made in the facilities of the Music Department at the University of Florida between
October and December 1998. A Sony (Sony Electronics, Inc., Park Ridge, NJ) TDC-8 DAT deck, an
ART professional tube mike preamplifier, and a Sony
WH20 head microphone were the equipment used.
The specifications of the equipment (microphones
and tape decks) used at both sites was such that dB
levels maintained were well within 3 dB with distortion levels less than 1% and the signal-to-noise ratio exceeding 50 dB within the analyzed frequency
range of 0.1 to 10 kHz.
Analysis procedures
Samples of forte production of the five cardinal
vowels sung at low, mid, and high pitch were digitized via Signalyze 3.1221 at a sampling rate of 44
kHz. A cross-section analysis of the harmonics of
each sung vowel was made at or near the center of the
steady-state portion of each vowel using analysis routines in Signalyze set to the narrow band setting. A
Power Macintosh 7600/132 was used to perform the
analysis. The values of each measurable harmonic detectable in the 2008000 Hz range was calculated in
respect to frequency (Hz) and amplitude (dB). This
information was then transferred to Microsoft Excel
5 22 and processed as a spreadsheet. Composite values
were then correlated for the singers in each group.
Discussion of subjects and procedures
The subjects ranged from 20 to 65 years in age. All
participants in this study were advanced singers who
considered their voices of stage quality with the accompanying appropriate resonance. All reported being in good health and in good singing voice at the
time of the recordings. In the recording sessions, each
participant was asked to produce their most resonant
voice, one they felt contained a singers formant and

SINGERS FORMANT IN SOPRANOS


would project on the stage. This usually yielded robust production and necessitated attenuation adjustments on the preamplifier. The existence of vibrato in
these highly trained voices caused the fundamental
pitch and harmonic array to shift considerably, and together with the live room acoustics, caused some
modulation effects. In both recording venues, the
rooms (basically practice rooms or studio practice
rooms) provided for a live acoustic environment.
RESULTS
General findings
Upon analysis it was found that the original Western Washington University results were, indeed, consistent with those obtained at the University of Florida. In spite of the different miking techniques and
subjects recorded at two totally separate sites, the end
results were markedly similar, attesting to the reliability of the experimental results. Basically, little evidence would support an FS similar to that of the

461

mens stage voices. With low-pitch and mid-pitch


vowels, broadband reinforcement (somewhat resembling the mens FS) was found in the 2.5 kHz and
higher range, but with a bandwidth generally 2 to 2.5
times greater than the equivalent mens FS. The reinforcement bands were manifested as envelope patterns of adjacent harmonics evident in a narrow band
cross-section display. With high-pitch vowels, there
was no obvious band of energy, but rather a general
strong distribution of partials extending in strong
voices to 58 kHz. There was also little distinction of
vowel quality associated with high-pitch production.
Specific findings
Although the different miking conditions yielded
different absolute dB values (affecting particularly
the higher harmonics) for the WWU and UFL
groups, the overall results were remarkably similar.
The specific findings are supported by spectrographic evidence and the data portayed in the composite
graphs in the figures cited.

FIGURE 1. High [e] production (921 Hz) showing weak harmonic production in expected
singers formant region.
Journal of Voice, Vol. 15, No. 4, 2001

462

RUDOLF WEISS ET AL

1. High-pitch vowels show no evidence of an FS .


The general curve shows that most harmonic
energy is concentrated in the F0 and diminishes
gradually in the higher harmonics. In fact, in
Figure 1, the weakest harmonic is shown to be
the one where one would expect reinforcement
for the FS. For a comparison of averages of
high to other pitch vowels, see Figures 2 and 3.
2. Mid-pitch vowels have the strongest FS type reinforcement. However, it is much broader than
the male FStypically from 2.6 to 4.6 kHz,
and generally at least 2 kHz in bandwidth,
compared to 500800 Hz found to be typical
for the male voice by Schutte and Miller1 and
our studies as well.5,6 See Figures 2 and 3.
3. Low-pitch vowels exhibited similar bandwidth
qualities to the mid-sung vowels. (See Figures
2 and 3.) However, there were varying degrees
of success among some singers in achieving a
ring at the lowest frequency.
4. When all low-pitch vowels are plotted, we find
a strong second formant influence for [e] and
[i]. This is evident in Figures 4 and 5 where all
vowel averages for both groups are plotted.

5. High-pitch vowels showed little differentiation


in vowel quality, even though accomplished
singers attempted techniques of formant tracking. There is a general decline of harmonic
strength evident particularly for the WWU
group, as seen in Figure 6, whereas the upper
harmonics are sustained at significantly higher
frequencies for the UFL group, as seen in Figure 7. This is attributable largely to the effect of
using a head microphone.
Other findings
1. The resonance peaks for low vowels, such as
they are, seem to be most vowel dependent and
are shifted progressively from low to high frequency in the succession [e a o u i] for the
WWU group in Figure 4, and in [e a u i o] for
the UFL group in Figure 5.
2. All mid and high-sung vowel productions have
the F0 appearing as the strongest partial with
subsequent partials decreasing in energy. (See
again Figures 2 and 3.) Group responses on individual vowels at mid-pitch level are indicated
in Figures 8 and 9. These can be compared to

WWU low/mid/high-pitch vowels


70

60

mid
50

low
high

dB

40

30

20

10

0
0

1000

2000

3000

4000

5000

6000

7000

8000

Hz

FIGURE 2. Averages of all harmonics for all vowels produced at three pitches by WWU group.
Journal of Voice, Vol. 15, No. 4, 2001

SINGERS FORMANT IN SOPRANOS

463

UFL low/mid/high pitch vowels


70

60

mid

50

low
40

dB

high

30

20

10

0
0

1000

2000

3000

4000

5000

6000

7000

8000

Hz

FIGURE 3. Averages of all harmonics for all vowels produced at three pitches by UFL group.

WWU low-pitch vowels


60

50
low [a]
low [e]

40

low [I]

dB

low [o]
low [u]

30

20

10

0
0

1000

2000

3000

4000

5000

6000

7000

8000

Hz

FIGURE 4. Harmonics of vowels at low pitch (261 Hz) for WWU group.

Journal of Voice, Vol. 15, No. 4, 2001

464

RUDOLF WEISS ET AL

UFL low-pitch vowels


60

50
[a]
[e]

40

[I]

dB

[o]
[u]

30

20

10

0
0

1000

2000

3000

4000

5000

6000

7000

8000

Hz

FIGURE 5. Harmonics of vowels at low pitch (261 Hz) for UFL group.

WWU High-pitch vowels


70

60
high [a]

50

high [e]
high [I]

40
dB

high [o]
high [u]

30

20

10

0
0

1000

2000

3000

4000

5000

6000

Hz

FIGURE 6. Harmonics of vowels at high pitch (932 Hz) for WWU group.

Journal of Voice, Vol. 15, No. 4, 2001

7000

8000

SINGERS FORMANT IN SOPRANOS

465

UFL high-pitch vowels


70

60
[a]

50

[e]
[I]
[o]

40

dB

[u]

30

20

10

0
0

1000

2000

3000

4000

5000

6000

7000

8000

Hz

FIGURE 7. Harmonics of vowels at high pitch (932 Hz) for UFL group.

WWU mid-pitch vowels


60

50

wwu - mid [a]


wwu - mid [e]
wwu - mid [I]

40

wwu - mid [o]

dB

wwu - mid [u]

30

20

10

0
0

1000

2000

3000

4000

5000

6000

7000

8000

Hz

FIGURE 8. Harmonic of vowels at mid pitch (622 Hz) for WWU group.

Journal of Voice, Vol. 15, No. 4, 2001

466

RUDOLF WEISS ET AL
UFL mid-pitch vowels
70

60
[a]

50

[e]
[I]
[o]

40

dB

[u]

30

20

10

0
0

1000

2000

3000

4000

5000

6000

7000

8000

Hz

FIGURE 9. Harmonics of vowels at mid pitch (622 Hz) for UFL group

the other pitches of individual vowel responses


in Figures 4, 5, 6, and 7.
3. All low-sung vowel productions show the F0 as
weaker in energy than the second and/or third
partial. This was true of all vowels except [i]
where in both groups the F0 was exceeded in
amplitude by the strongest F2 harmonic.
4. Weaker voices tended to display a nearly sine
wave production on high pitch vowels; strong
voices much less so.
5. Some voices were so strong that even upon
close unidirectional miking, clear saturation interference patterns were recorded. This resulted
in an almost square wave production, an anomaly with one unusually strong singer. Generally
these effects diminished when the amplitude of
production was decreased.
DISCUSSION
This study supported, for the most part, our previous findings 6 and those of Sundberg,7 and added
specific data regarding the nature of the sopranos
projected voice. The spectral analysis procedure, plotting the frequency and dB value of each visible harJournal of Voice, Vol. 15, No. 4, 2001

monic, proved to be a superior technique (for this


type of study) over the linear predictive coding
(LPC) method and broadband spectrogram approach
used in our earlier study.6 The Signalyze program21
and the analysis routines worked well and gave consistent readings.
Although the actual frequencies of each harmonic
and fundamental frequency were plotted and averaged, the absolute values were not considered significant to this study. As expected, the fundamental frequencies and, thus, the frequencies of the harmonic
multiples varied considerably, especially for the highpitch vowels. Both the individual efforts and the effect of the vibrato caused values for the high fundamental frequencies to vary as much as 25 Hz. No
effort was made to avoid the issue of the vibrato since
it is a natural manifestation of trained voice production.
Even though all voices were chosen at the recommendation of the respective trainers for their strength,
some voices were, impressionistically at least, considerably stronger than others. The strongest voices
tended to have stronger upper harmonics with less
loss of amplitude at frequencies over 5 kHz. Their
production tended to emulate square wave configura-

SINGERS FORMANT IN SOPRANOS


tions. Weaker voices tended more toward sine wave
productions. The square wave productions may also
have been influenced by live room acoustics and
wave reflection activity resulting from extremely
powerful voices in a relatively small enclosure.
The use of two groups was deemed both effective
and productive. It provided not only a bigger sampling body, but allowed clarification of issues raised
in the first study.6 Although all the tested voices
would likely carry well in a concert setting, especially at the high-pitch forte production, the strongest individual voices were found in the second group. Testing similar voices in large halls might also prove
interesting, but it is doubtful that significantly different results would be attained. Low-pitch production
for sopranos seemed clearly more problematic for
projection purposes than did mid-pitch and high-pitch
efforts. Certainly, the 261-Hz productions requested
of the sopranos were not in their optimal range.

467

appropriate for sopranos to encourage a voice facilitating maximal projection. Vowel differentiation is
unrealistic and virtually unattainable when attempting extremely high notes, and high-frequency reinforcement considerably broader than that of the male
FS is a common denominator among strong female
voices even at mid-pitch and low-pitch production.
The broadband high harmonic reinforcement phenomenon typical of strongly projecting soprano voices could simply be called extended frequency reinforcement or EFR for want of a better term to
distinguish this from the FS.
Acknowledgements: This study was funded in part by a
grant from the Bureau for Faculty Research, Western
Washington University. Thanks is also extended to the Institute for the Advanced Study of the Communication
Processes, University of Florida, for the generous use of
their facilities, to the willing participants and their voice
coaches, and to Donna Mustard and Caite York for their
valuable assistance in the preparation of the article.

CONCLUSION
This study, performed in two stages with two sets
of advanced soprano singers, yielded consistent results showing that there is strong resonance in a
broad spectrum from approximately 2.6 to 4.6 kHz
for low-sung vowels. For high-sung vowels, there
was no typical band of reinforcement. The first four
or five harmonics (to about 4.7 kHz) were typically
reinforced. Weaker voices tended to be distinguished
by harmonics becoming progressively weaker at
higher frequency levels, sometimes so weak as not to
register at all; the strongest voices were able to extend their harmonic strength well into a higher part of
the spectrum. In some cases a range of reinforcement
extended considerably further upward and strong energy could still be detected in the 810 kHz range.
Furthermore, despite efforts of formant tracking,
vowel quality differentiation at high-pitch levels is
almost negligible.
Since the wider reinforcement band of 2 kHz or
more for the soprano voice at low-pitch and midpitch levels were considerably broader than that of
the mens FS, it could be safely assumed that the
physiological manifestations necessary to produce
this effect are also dissimilar to the mens (which
usually involves lowering the larynx). This would
imply that different pedagogical techniques might be

REFERENCES
1. Schutte HK, Miller R. Intraindividual parameters of the
singers formant. Folia Phoniatr. 1985;37:3135.
2. Bartholomew WT. A physical definition of good voice
quality in the male voice. JASA. 1934;6:2533.
3. Sundberg J. Articulatory interpretation of the singing formant. JASA. 1974;55:838844.
4. Coffin B. The relationship of phonation and resonation.
NATS Bull. 1975;FebMarch:3744.
5. Morris J, Weiss R. The singers formant revisited: pedagogical implications based on a new study. J Singing. 1997;
53(3):2125.
6. Weiss R, Morris J. The Correlation of The Singers Formant
as a Factor of Dynamics, Pitch and Vowel Variance: Trained
vs. Non-Trained Singers. In: Zeitschrift fr Dialektologie
und Linguistik. Supplement 96: Untersuchungen zu Stimme
und Sprache. Steiner Verlag; Stuttgart, Germany: 1996:5474.
7. Sundberg J. Vocal tract resonance in singing. The NATS
Journal. 1988;MarchApril:1131
8. Schultz-Coulon H-J, Battmer R-D, Riechers H. Der 3-kHzFormantein Ma fr die Tragfhigkeit der Stimme? II.
Die trainierte Singstimme. Folia Phoniatr. 1979:31;302
313.
9. Titze IR. Male-female differences in the larynx. The NATS
Journal. 1988, JanFeb:31.
10. Winckel F. How to measure the effectiveness of stage
singers voices. Folia Phoniatr. 1971:23;228233.
11. Nawka T, Anders LC, Cebulla M, Zurakowski D. The
speakers formant in male voices. J Voice. 1997:11(4);
422428.
Journal of Voice, Vol. 15, No. 4, 2001

468

RUDOLF WEISS ET AL

12. Sundberg J. Swedish Voices in Music. Proceedings of the


XIIth ICPhS. Stockholm University; Stockholm, Sweden:
1995;1:28.
13. Sengupta R. Study on some aspect of the singers formant
in North Indian classical singing. J Voice. 1990:4(20);
12934.
14. Cleveland TF. A closer look at formant tracking. NATS.
1994:JanFeb:4143.
15. Sundberg J. The singers formant revisited. Speech Transmission Lab: Quarterly Programs and Stats Report. 1995:
Oct. 15:8396.
16. Titze IR. Subglottal resonances. The NATS Journal. 1988:
MarchApril;2829.

Journal of Voice, Vol. 15, No. 4, 2001

17. Titze IR, Story BH. Acoustic interactions of the voice source
with the lower vocal tract. JASA. 1997:101(4);22343.
18. Titze IR. The wide pharynx. J Singing. 1998:55(1);2728.
19. Timberlake, C. Pedagogical perspectives, past and present:
laryngeal positioning. The NATS Journal. 1994:SeptOct.;
3942.
20. Hollien H. That golden voicetalent or training? J Voice.
1993:7(3);195205.
21. Signalyze Version 3.12; Eric Keller, InfoSignal Inc., C.P 72,
CH-1015, Lausanne, Switzerland;1997.
22. Excel 98. Copyright 19851998, Microsoft Corporation;
1998. Microsoft Corporation, One Microsoft Way, Redmond, WA 98502-6399.

Das könnte Ihnen auch gefallen