VOICE RESEARCH AND TECHNOLOGY

Ingo Titze, Associate Editor

Resurrection from the Coffin

Ingo R. Titze

fl
he first edition of Berton Coffin's book onthe sounds of singing ap-
peared over thirty years ago, followed by a second edition about ten
years later.' Coffins work was deeply rooted in the linear source-
T filter theory of vowels, which in his day was the workhorse of speech
- science. 1-us primary colleague in voice science was Pierre Delattre, a phoneti-
cian-acoustician who made major contributions to vowel theory as it was
evolving in Europe, Japan, and the United States in the second half of the twen-
tieth century.
The pedagogic element in Coffin's work was the "favorable vowel chart:'
Ingo R. Ii tie
later published in color as the "chromatic vowel chart for voice building and
tone placing:' Based on the then accepted linear source-filter theory of vow-
els, the chart shows which vowels most likely resonate the fundamental fre-
quency and harmonics of any note in the vocal range. The chart is compre-
hensive and complete, but somewhat antiquated now because the most
interesting pitch-vowel interaction phenomena are based on a nonlinear
source-filter theory. Thus, the old chart is in the coffin, but the Coffin idea needs
to be resurrected. This little essay is the beginning of such a resurrection,
with, one hopes, many additional installments to come.
The main difference between the current nonlinear source-filter interac-
tion theory and the older linear theory is that formants need to he avoided,
not sought out, in the choice of a vowel. It is still true that formants (the res-
onances of the vocal tract) can boost the energy of any harmonic of the source,
but this selective "boosting" often creates irregularity in the vibration pat-
tern of the vocal folds. Unlike in a woodwind or brass instrument, where the
horn is long and steady in its geometry, the vocal tract in humans is relatively
short and constantly changing due to articulation of phonemes. This does
not permit an orderly line-tip of source harmonics with vocal tract formants.
Hence, it is better to keep the harmonics away from the formants. Furthermore,
nonlinear source-filter theory predicts a preference in placing harmonics on
the left side of a formant (below the formant frequency) as opposed to the
right side (above the forniant frequency). Fortunately, there is more than one
formant, so vowels can be chosen such that a "leap over the formant" can be
made by a harmonic so that it lands on the upslope of an adjacent formant.
Thus, chasing "favorable vowels:' as Coffin called them, on an ascending
pitch scale is like walking up a tall mountain that has multiple peaks and val-
Journal of Singing, November/ December 2007 leys along the way. You stay on the upslopes and try to leap over the valleys
Volume 64, No. 2. pp. 199-201 . -- . . . . .
Copyright 2007 as quickly and effectively as possible. The difficulty is that if you are the
National Association of Teachers of Singing harmonic, all your family members (the other harmonics) have to do the

NovEiBER/DEcEslBER 2007 199



Ingo R. Titze

tract inertance. Vocal tract inertance has been shown to

Fo 2Fo 3Fo 4Fo
enhance vocal fold vibration. 2 The greater the inertance,
the more enhancement there will be for a harmonic of
the source. As seen in the figure, inertance changes rap-
uh
idly onl y near the formants. For the ii = lit vowel, for
aw
example, the first formant is near C 4 ; for the ih II /
vowel, the first formant is near Cn; for the ae = / x / vowel,
Ge
the first formant is between C:, and C 6 (note the pitch
labels C2, C, C41 C5, and C6 on horizontal axes). For any
selected pitch, there will be a series of harmonics cre-
le
ated at the source. The first four harmonics are shown as
vertical lines in Figure 1 for a pitch C 4 . Note the loca-
C2 D T5 C6 '
tion of these harmonics in the peak and valley terrain
IT 10
Frequency in Hz of the vowels. Stable and acoustically strong productions
result when each of these harmonics (F 0, 2F0, 3F0, and
Figure 1. New Coffin chart.
417,, as labeled on top) resides in high inertance terri-
same thing in lock-step. You walk in a row, separated by tory, which is to the left of a formant. Whenever a dom-
a constant distance, but the peaks and valleys are not inant harmonic lands in the formant region, where in-
equally apart. ertance rises and falls quickly, vocal fold vibration can be
Figure 1 shows the process. Ten vowels are represented destabilized, or the sound output changes suddenly from
as nearly horizontal lines. The double letter vowel labels strong to weak (or vice versa). For the C 4 pitch chosen
have the following phonetic interpretation: uu = /u I, 00 in our illustration, F is in trouble with the vowel ii = Ii!,
= lol, uh = IUI, aw = lI, aa = Ia!, ae = lx!, eh = l/, ih 2F(J is in trouble with the vowel ih = /1/, and 3F0 is in
=11/, ee = Ic!, and ii = lit. The numbers on the vertical trouble with the vowels aa = lot and ae l/. As the
axis are an arbitrary scale for vowel separation. The peaks pitch changes, of course, different vowels present prob-
and valleys in the curves represent the changes in vocal lems for the harmonics. Thus, the singer modifies the

PUBLICATION DEADLINES
Vol. 64 No.1 Vol. 64 No.2 Vol. 64 No.3 Vol. 64 No.4 Vol. 64 No.5
Sept/Oct 2007 Nov/Dec 2007 Jan/Feb 2008 Mar/Apr 2008 May/June 2008

Article manuscripts should be submitted directly to the Editor-in-Chief All manuscripts are evaluated by the Editorial Board.
Decision to publish rests with the Editor-in-Chief. The deadlines published here refer to articles that have already been accepted
for publication.

Features and Reviews by Continuing Contributors

01 May 2007 01 July 2007 01 Sept 2007 01 Nov 2007 01 Jan 2008
Submission of Advertisements
01 July 2007 01 Sept 2007 01 Nov 2007 01 Jan 2008 01 Mar 2008
Issue to be placed in mail approximately
03 Sept 2007 03 Nov 2007 04 Jan 2008 05 Mar 2008 05 May 2008
Issue to be received by subscribers approximately
17 Sept 2007 17 Nov 2007 18 Jan 2008 19 Mar 2008 19 May 2008
(Dates of delivery may vary depending upon local office policies.)

200 JOURNAl. OF SINGING

 Voice Research and Technology

vowels to maintain consistency of harmonic energy Laryngological Association Award. He is a Fellow of the Acoustical Sod-
throughout the pitch range. ety of America and the American Speech-Language-Hearing Association.
How is this different from what Coffin taught? In Dr. Titze has served on a number of national advisory boards and scien-
tific review groups, including the Scientific Advisory Board of the Voice
principle, there is no difference. There are good vowels
Foundation and the Division of Research Grants of the National Institutes
and bad vowels for a given pitch. In practice, however,
of Health. In addition to his scientific endeavors, Dr. Tit2e continues to be
the new theory is less restrictive because harmonics and
active as a singer. He is married to Kathy Titze and has four children.
formants do not need to be fine-tuned to each other. It Mail should be addressed to Ingo R. Titze, National Center for Voice and
is only important to "lift" a harmonic over an inertance Speech, 330 WJSHC, Iowa City, IA 52242. Telephone (319) 335-6600.
valley by modifying the vowel. The new theory also
makes a strong connection between vowel modification
and voice registers (sometimes called lifts), a topic that N1vIioN1I. AssocI,&rIoN OF TEACHERS OF SINGING
will be addressed in later essays. The current message
Membership:
is simply that Coffin is not in the coffin, but resurrected
Membership in the National Association o fii achers
via a new vowel chart. The final (immortal) version of this
of Singing is open to an y citizen of any country whose
chart is yet to be revealed.
professional training and experience qualifying him
or her as a teacher of singing.
NOTES
Associatiate Membership is available for voice
I. Berton Coffin, Coffin's Sounds of Singing: Principles and teachers and advanced students who have not as yet
Applications of Vocal Techniques with Chromatic Vowel Chart, completed the requirements for full membership.
2" Edition (Metuchen, NJ: The Scarecrow Press, 1987).
Affiliate Membership is open to persons or groups
2. Ingo R. Titze. "The Physics of Small-Amplitude Oscillation that are interested in vocal pursuits but are not actually
of the Vocal Folds,' Journal of the Acoustical Society ofAmerica involved in the teaching of singing, such as speech
83, no. 4(1988): 1536-1552.
therapists, laryngologists, schools, publishers, and
music stores.

Ingo R. Titze is Distinguished Professor of Speech Science and Voice at Publications:

the University of Iowa and Executive Director of the National Center for Membership includes a subscription to the 7ouriialof
Voice and Speech at the Denver Center for the Performing Arts. His for- Singing, the official journal of NATS, and to inter
mal education is in physics and electrical engineering, but he has devoted Nos, the NATS Newsletter.
much of his studies to vocal music and speech. Dr. Titze has published
more than 500 articles in scientific and educational journals, coedited Information:

two books titled Vocal Fold Physiology, and has authored two books Application for Membership forms ma y be secured
called Principles of Voice Production, and The Myoelastic Aerodynamic from the Executive Director, NATS, 9957 Moorings
Theory of Plionotion. He has lectured throughout the world and has ap- Drive, #401, Jacksonville, FL 32257. These forms
peared on such educational television series as Innovation, Quantum, contain more detailed information about the qualifica-
and Beyond 2000. He is a recipient of the William and Harriott Gould tions for membership and the Code of Ethics of the
Award for laryngeal physiology, the Jacob Javits Neuroscience Investigation National Association of Teachers and Singing.
Award, the Claude Pepper Award, the Quintana Award, and the American

NOVEMBER/DECEMBER 2007 201