{IHHIWIHIIWWIHJIWIHIIHIWUWIIHJHNHHI

       

(J14
I\]_\

   

'—I

Wuhan-S

NEIV ERS ITYLIBRARI

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII ‘IIIIIIIIII IIIIIIII

3 1293 00881 280

 

 

 

 

 

 

 

 

 

 

This is to certify that the

thesis entitled
Measurement of Fundamental Frequency via

Acoustic and Inductive MicrOphones

presented by

Holly Beth Hochroth

has been accepted towards fulfillment
of the requirements for

Master of Arts degree in §peech-Language
Pathology

Major professor

 

Date Q’éz/ 7‘2 /?.93

/

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

 

 

LIBRARY
Michigan State
University

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.

DATE DUE DATE DUE DATE DUE I

[:1
I”?
L__I

    
 
  

 
 

 

 

 

 

L
I:

 

 

 

_III
II

 

 

 

 

 

 

 

 

L
D

 

 

 

 

 

 

 

___I
I

 

 

 

 

 

 

 

 

 

 

 

 

MSU is An Affirmative Action/Equal Opportunity institution
«=sz

 

MEASUREMENT OF FUNDAMENTAL FREQUENCY
VIA ACOUSTIC AND INDUCTIVE MICROPHONES

By

Holly Beth Hochroth

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degee of
MASTER OF ARTS
Audiology and Speech Sciences

1993

 

ABSTRACT

MEASURMENT OF FUNDAMENTAL FREQUENCY
VIA ACOUSTIC AND INDUCTIVE MICROPHONES

By
Holly Beth Hochroth

The purpose of this study was to determine whether measurements of fundamental
frequency derived from 2 different microphones were valid when compared to accepted
acoustic measurements. This study involved 25 normal subjects. Acoustic data were
collected via the hand-held Visi-Pitch microphone placed directly in front of the mouth.
Frequency measurements derived from stroboscopic instrumentation were obtained via an
inductive microphone placed on the surface of the neck on the thyroid lamina of the
larynx. Both microphones were utilized simultaneously to elicit the same segment of
voicing for the vowels /a/ and /i/. A strong positive correlation was found to exist between
the frequency measurements obtained. The frequency differences between the
measurements with each microphone indicated a statistically significant difference.
Frequency differences obtained were < 1.0 Hz. The measurement of fundamental
frequency of steady state vowels with the acoustic and inductive microphones were found

to be clinically comparable.

dedicated to
Mark Scott Kluger

for being patient through the long nights

 

ACKNOWLEDGMENTS

My heartfelt appreciation goes to Dr. Peter LaPine, for spending so much time and effort
in helping reach my goals; to Dr. Paul Cooke and Dr. Leo Deal for being on my thesis
committee and for being so patient; to Ms. Jan Lewin, for allowing me to complete my

research when I thought no one else would.

I want to thank my parents for allowing me the opportunity to explore all my options and
to pursue all of my dreams. I could not have done this without both of you. To Mark,
thanks for being there when I needed you.

LIST OF TABLES.

TABLE OF CONTENTS

LIST OF FIGURES. .

DEA—FEE

I

Introduction

Introduction. .
Statement of the Problem.
Purpose.

Research Questions.
Hypothesis.

Review of the Literature.

History of the Measurement of
Fundamental Frequency.

Present Instrumentation

History of Stroboscopy.

Fundamental Frequencies of Vowels. .

Methodology. .

Subject Selection.

Data Collection. .
Procedures and Instrumentation.
Data Analysis. .

Statistical Analysis.

Results.

Results.

unguar—

17

17
18
18
20
21

t»)
to

 

V Discussion, Summary and Conclusions.
Introduction.
Summary and Discussion.
Conclusions.
GLOSSARY .
APPENDICES
A Averages for each subjects vowel productions.
B Mean Frequencies for 3 trials. .
C Norms for Speaking Fundamental Frequency.
D Case History Checklist. .
E Informed Consent.
LIST OF REFERENCES.

BIBLIOGRAPHY.

vi

33

33
34
36

38
42
44
46
47
48
49

52

LIST OF TABLES

Mean Frequency Measurements for Males, Females,
and entire sample

Mean Differences of the Calculated Differences
for 3 trials for each subject

t-values for /a/ and W for entire sample
t-values for sample size N=24 .
Ranges for Mean Frequencies.

Measures of Central Tendency for entire sample.

2.392

23

26

27

30

31

 

la

lb

LIST OF FIGURES

Mean Frequencies for /'a’
Mean Frequencies for /i/

Mean Frequency Differences between Microphones

EAQE

24

 

Chapter I

Introduction

Wen

Voice is usually the first audible indication of a person's gender, personality, health
and even emotional state. A "shaky" voice can indicate nervousness, tension or fear. A
speaker with a strong and clear voice may be perceived as having strength, confidence and
security. Voice allows the listener to distinguish a person's gender and age. Children at
young ages typically have high pitched voices; however, a high pitched voice similar to
that of a child is inappropriate when produced by an adult (Colton and Casper, 1990). As
a person ages, voice changes. Ifa large man has a very high pitched voice, similar to that
of a woman or a child, his character may be judged as feminine, boyish or weak. First
impressions are produced as soon as a person speaks; therefore, voice is an important part
of an individual’s personality.

Voice is also the first audible indication that there might be a physical problem in
the larynx. Zemlin (1988) provides a description of the function of the larynx:

The larynx is a modification of the uppermost tracheal cartilages. It forms a highly
specialized valvular mechanism that may open or close the air passageway. An
extremely important function of the larynx is to serve as a protective device. (p. 35)

Stemple (1984) described the larynx as "A tube with several folds. It is composed of
ligaments, muscles, and mucous membranes" (p.13). The vocal folds are located within
the larynx, and are located superior to the trachea, or wind pipe. Expired air from the
trachea and lungs vibrates the vocal folds and produces sound.

The morphological structure of a vocal fold is described by Hirano (1974) as
having five discrete layers: 1) the epithelium, which helps maintain the shape of the vocal
fold; 2) the superficial layer of the lamina propria, also known as Reinke's space; 3) the
intermediate layer of the lamina propria, consisting of elastic fibers that act like soft rubber

2 Measurement of fundamental...

bands; 4) the deep layer of the lamina propria, made up mostly of collagenous fibers; and
5) the vocalis muscle, the main body of the vocal fold (Zemlin, 1988). The structural
properties of the vocal folds allows for elasticin to change during voice production. The
vocal folds have the property of mass which varies for different persons, varies as a
function of fundamental frequency, varies during the vibratory cycle, or varies when
augmented by pathological growth or disease (Colton, 1988). When there is a change in
the morphological structure of the vocal fold(s), normal production of voice is altered. A
polyp, for example, adds mass to the vocal folds, thereby changing the normal vibration of
the folds. As a result, a lower pitch and a change in vocal quality are produced. When
voice is produced, it is typically perceived by a listener with respect to its quality, loudness
and pitch (Colton and Casper, 1990). When a person speaks, these broad perceptual
categories can be described by a listener. Adjectives are a necessary component in
describing what a person‘s voice sounds like. Therefore, voice may be evaluated and
described clinically using adjectives from these three categories.

Colton and Casper (1990) suggested the following eight terms to define the
subjective parameters of quality: hoarseness and roughness, breathiness, tension, tremor,
strain/struggle behavior, sudden interruption of voicing, and diplophonia. These difiering
vocal qualities can indicate various voice disorders. For example, a hoarse voice may result
after shouting all day at an exciting football game. Physical changes like edema, or
specifically swelling of the vocal folds, may be the culprit in this case. Any physical change
to the vocal folds will afiect the resulting vocal quality. The additional fluids present as
swelling occurs makes normal vibration difiicult. I

Loudness is the psychological sensation that can be measured as intensity. Intensity
is the physical property of an acoustic signal, and the loudness of that signal is related to its
intensity (Borden and Harris, 1984). Loudness is a subjective measure because the ear
does not hear all sounds with the same sensitivity (Martin, 1986). It is often necessary to

adjust the loudness level of voice in various situations. For example, the loudness level of

3 Measurement of fundamental...

voice that is appropriate in a library is not apprOpriate in public speaking. Speech loudness
level is regulated by the surroundings and auditory feedback in a person with normal
hearing.

Pitch is the psychological perception of frequency of vibration. Pitch can be
described subjectively by a listener as being low or high, or audible. Although perception
of quality, loudness and pitch perception are the first components in determining a voice
disorder, viewing the vocal folds and the vocal tract allow for a more thorough diagnosis of
the voice.

S te ent o e oblem

Many different methods are available to evaluate and diagnose voice disorders.
Subjective diagnostic methods are considered reliable but are not always considered valid.
It is becoming increasingly important to have objective and valid methods with which
professionals can measure voice and diagnose causal relationships. It is necessary to
compare different measurements obtained by new diagnostic tools to determine validity and
relevance for a thorough diagnosis and a subsequent therapy plan, if appropriate.

Some technology available today allows for viewing the vocal folds at both a
normal rate of vocal fold vibration and at a seemingly slower rate able to extract a
simultaneous measure of frequency. Measurement of voice production can be obtained
with different types of endoscopes in combination with an instrument known as a

synchronstroboscope. Boone and McFarlane (1988) described an endoscope as:

An instrument that is introduced intraorally or intranasally; the light on the tip of the
scope (which comes fiberoptically from an external light source) illuminates the
nasal and oral pharynx that is Viewed through a window lens on the tip of the
endoscope. When the oral or nasal endoscope is attached to a small video camera,
the examination is termed a videoendoscopic evaluation. When using a stroboscopic
light source instead of a steady state light source, a slow motion observation of the
vocal physiology is produced. (p.91-92)

4 Measurement of fundamental...

Rigid oral endoscopes and flexible fiberoptic nasopharyngoscopes are used commonly
today. Both the oral and nasal endoscopes are illuminated by an external light source.
Watterson and McFarlane (1991) referred to these light sources as ”cold lights" because
heat is not transferred through the endoscope but is dispersed externally. Another light
source used is the strobe light. Watterson and McFarlane (1991) stated that "Stroboscopic
light sources illuminate the vocal folds at intervals corresponding to vocal frequency. Thus,
while high-speed photography samples successive points within each vibratory cycle,
stroboscopy samples successive points across vibratory cycles" (p.81). This light source
allows the examiner to view the vocal folds and make determinations regarding structure,
symmetry, and mucosal wave. The frequency of the strobe light is generated by the
fundamental frequency that is produced. This is accomplished via the inductive
microphone, connected by a cable to the synchronstroboscope, that is held against the neck
at the level of the thyroid lamina of the larynx. In discussing fundamental frequency (F 0),
it is important to keep in mind the source of the sound. Colton (1988) discussed the three
physical properties of the vocal folds that affect fundamental frequency: tension, mass,
and length. He stated that control of the vibrating frequency of the vocal folds is caused by
manipulating one or more of these 3 properties. For each individual, the manipulation
needed to change these properties of the vocal folds is different. There are differences in
these 3 variables between men and women, adults and children. These differences are
demonstrated in the pitches produced by men and women. By lengthening the vocal folds,
pitch increases because of the decreased mass and increased tightness of the vocal folds.
This lengthening allows the vocal folds to vibrate at a faster rate. Mass is the thickness of
the vocal folds. The thinner and tighter the vocal folds, the higher the resulting pitch will
be. Conversely, the thicker and more lax the folds, the lower the resulting pitch. The
tension of the vocal ligament is regulated through the rocking of the arytenoid cartilages
along the cricoarytenoid joints. Similar to a rubber band, when the vocal ligament is pulled

tightly, it will vibrate at an increased rate. When lax, the rate of vibration is decreased.

 

5 Measurement of fundamental. ..

The rate of vibration of the vocal folds is frequency, that is, the number of the
cycles at which the vocal folds vibrate; and the acoustical and psychological perception of
frequency can be called pitch. Colton and Casper (1990) defined the term fundamental
frequency as "The most frequently or commonly occurring frequency that characterizes a
particular vocal production" (p. 184). According to the definition of fimdamental
frequency by Colton and Casper (1990), during a single production of a steady state vowel,
the frequency that occurs most often is considered the fundamental. The other frequencies
that are often present are the overtones, or harmonics. Fundamental frequency is the
lowest component of a complex tone (Borden and Harris, 1984). Sustained vowel
phonation and conversational speech are comprised of complex tones. The frequencies
produced at the level of the larynx are changed as they travel through the vocal tract.
When measuring the fundamental frequency of a complex tone, such as speech, it is the
lowest component of the tones that are produced that is considered the fundamental. The
unit of measurement for fundamental frequency is cycles per second (cps). However, in
recent years, the term hertz (Hz) has been adopted instead of cps in honor of the
nineteenth century German physicist Heinrich Hertz" (Martin, 1986, p.21). The terms
”Hertz" and "cps" can be used interchangeably.

Zemlin (1988) described the vibratory cycle in terms of phases: the opening phase,
the closing phase and the closed phase. The vibratory cycle is periodic during normal
phonation. Borden and Harris (1984) discuss that, "During sustained vowels, for example,
the folds open and close in a certain pattern of movement which nearly repeats itself. This
action produces a barrage of air bursts that set up an audible pressure wave (sound) at the
glottis” (p. 84). The Myoelastic Aerodynamic theory of vocal fold vibration explains how

the vocal folds are set in motion to create the vibratory cycle and phonation:

This theory suggests that the vocal folds are set into vibration by interaction of
subglottic air pressure and the passively approximated vocal folds. In short, the
vocal folds adduct and approximate at the midline. Subglottic air pressure fi'om the
lungs rises against the resistance of the closed folds. This pressure eventually

 

6 Measurement of fundamental...

overcomes the resistance, forcing the folds apart and releasing a small puff of air.
This release of air creates a sudden drop in air pressure at the level of the vocal
folds, creating a suction. This suction, called the Bernoulli effect, along with the
static positioning of the adducted folds, begin to draw the folds back together. The
closer the folds draw, the greater the suction, until the folds snap shut, completing
one vibratory cycle. (Stemple, 1984, p. 46)

Increases in subglottic air pressure work to change both the frequency level and the
intensity. The vocal folds are pulled together with increased force, and increased force is
therefore necessary to blow the folds apart to begin phonation. Frequency will increase if
the vocal folds are tense; intensity will increase with the amount of subglottic air and force
necessary to blow the folds apart. The ability to increase or decrease subglottic air pressure
allows a speaker to change vocal pitch to fit a situation appropriately.

There are differences in the ranges of fundamental frequencies produced by the
human voice. Men may have a speaking fundamental frequency range that is typically
between 80 and 150 Hz, whereas women's fundamental frequencies may range anywhere
between 150 and 250 Hz. However, it is not uncommon for any individuals speaking
fundamental frequency to be outside of these particular ranges. This difference in
frequency range is largely due to the physical size differences between the larynges of men
and women. The frequencies that are present in a periodic complex sound, known as

harmonics, are described by Martin (1986) as:

whole number multiples of the fundamental. These tones, which occur over the
fundamental, are called harmonics or overtones. With respect to periodic signals,
the only difference between overtones and harmonics is the way in which they are
numbered: the first harmonic is the fundamental frequency, the second is twice the
fundamental, and so on. The first overtone is equal to the second harmonic, and
further overtones are numbered consecutively. (p.35)

The parameter to be examined in this research is fundamental frequency (F0).

Different conditions can cause fundamental frequency to change. Benign lesions, such as a

‘7 Measurement of fundamental...

nodule or vocal polyp, lower pitch and quality and voice is thereby perceived by a listener
as hoarse and breathy. Edema, or swelling of the vocal folds, can also cause the
fundamental frequency to decrease. Too much tension in the larynx can cause cracking
and breaking of the voice; laryngitis may result. Different positioning of the larynx during
laryngeal examinations can also affect voice production. By repositioning the larynx, the
vocal tract is altered, and sound production varies. It can be said that the human voice has
variability from day to day depending on a person’s mood, speaking context, and general
health.

Purpose

The purpose of this research is to determine whether the measurement of the
fundamental frequency for the vowels /a/ and /i/ with the Visi-Pitch microphone and the
Stroboscope inductive microphone differ in a systematic way. A second purpose is to
determine if the frequency measurements obtained with these microphones and instruments

are reliable?

Resegch Questions
This research will examine the following questions:

1) Does the measurement of frequency differ for the vowels /a/ and iii in a systematic way
using the Visi-Pitch acoustic microphone and the stroboscope inductive microphone?
2) Are the frequency measurements obtained with the Visi-Pitch acoustic microphone and

the stroboscope inductive microphone valid?

Hypothesis

There is no difference in the fundamental frequency measurements for the vowels
W and /i/ when measured using the synchronstroboscope and an inductive microphone and

the Visi-Pitch with the acoustic microphone.

Chapter II

Review of the Literature

Histog; of the Measurement of Fundamental Frequency

Fundamental frequency, the acoustic correlate of "pitch,“ is measured using a
variety of different methods and techniques. Baken (1987) described how past methods of
fundamental frequency measurement were diflicult, time consuming and expensive. The
earlier methods did not work with electronic circuits. Baken (1987) described methods
including the phonodeik by Anderson in 1925 which "used light reflections from a mirror
on a wire made to vibrate by collecting sound power with a diaphragm to produce curves
on a photographic film“ (p. 128). Baken further described other methods of determining
fundamental frequency by measuring graphic readouts and counting tape striations from a
recorded sample. One example of this is the reading of a spectrogram.

The spectrograph was developed in the 1940s at Bell Laboratories by Ralph Potter
(Borden and Harris, 1984). This instrument “allowed investigators to conveniently analyze
the frequencies represented in speech across time, producing a visual display called a
spectrogram“ (Borden and Harris, 1984, p.21). Baken (1987) explains that although
sound spectrography is a powerful tool, "The interpretation of spectrograms rests on a
thorough knowledge of the acoustics and physiology of speech, on familiarity with the way
in which spectrograms are generated, on experience, and on a trained eye" (p.316). Sound
spectrography can be difficult and time-consuming for the examiner, and interpretation of
the data may be diflicult. I

Another inductive method to determine fundamental frequency is the Fundamental
Frequency Indicator. A microphone is held against the larynx or placed under the nose
while producing a sustained vowel or the consonant /m/. Pitch levels are read from the

dial, and the frequency bands can be set to indicate when a certain pitch is achieved

 

9 Measurement of fundamental...

(Boone and McFarlane, 1988). This instrument is useful in facilitating a new pitch level in
voice patients.

There are also other noninstrurnental measurements of fundamental frequency that
are judged to be valid. Boone and McFarlane (1988) state that "It is possible to measure
frequency range and make other measures of frequency without instrumentation other than
a piano or a pitch pipe. An initial voice recording made at the time of a patient's first clinic E
visit is a useful tool for analyzing the patient's habitual pitch level. This analysis can be
made after the patient has left the clinic. One method we have used is to stop the recorder
at random points and attempt to match the voice pitch level with a pitch pipe or a piano.

After some experience with a pitch pipe, it is possible to match pitch levels between the

 

pitch pipe frequency and the patient's voice" (p. 103). These methods require perceptual
judgments to be made by the examiner. One important aspect that is measured in this way
is the concept of habitual pitch. Coleman and Markham (1991) define habitual pitch as a
"Single pitch or narrow range of pitches that the individual uses most of the time" (p.173).
Habitual pitch extracted as the fundamental frequency is the parameter to be measured in
the present study.
Present Instrumentation

The Visi-Pitch (Kay Elemetrics) is an instrument that can be used to measure
fundamental frequency. Kamell (1991) states that "The Kay Elemetrics Visi-Pitch has
become increasingly popular as a research tool for measuring fundamental frequency and
pitch perturbation" (p.91). When the Visi-Pitch 6097 or 6098 are used in combination
with a computer interface, the instruments offers features that the Visi-Pitch 6087DS

alone does not. Kamell (1991) describe the functions of the computer interface:

The sole function of the microcomputer is to perform the frequency, amplitude,
and perturbation calculations and to display the graphics. All of the waveform
processing, pitch extraction, and measurement processes are completed using the
undigitized, or analog, signal. This is a major advantage of the Visi-Pitch,
particularly for clinical applications, because analog processing is faster than digital

10 Measurement of fundamental...

processing. Consequently, fundamental frequency data can be calculated and
plotted on the computer screen almost as quickly as it is input to the device (p.91 ).

The computer interface allows the examiner to save time by calculating the statistics during
voice production. Horii (1983) described the Visi-Pitch as a "...portable instrument that
extracts and displays F o in real time, either from a live voice via a microphone or from a
tape recording" (p.468). A hand-held microphone is connected to the unit, and a
measurement of fundamental frequency can be taken alone or in combination with
intensity. The microphone used with a Visi-Pitch to obtain voice recordings is an acoustic
microphone. These types of microphones are responsive to sound pressure wave variation
and convert the pressure variations into time-varying electrical signals (Decker, 1990).
These types of microphones can be uni- or multi-directional. Decker (1990) stated that a
higher signal-to-noise ratio is achieved when a unidirectional microphone is placed very
close to and directly in front of the mouth. It is important to have a microphone that is not
sensitive to ambient noise. Holding the acoustic microphone in close proximity to the lips
facilitates the higher signal-to-noise ratio and the microphone does not pick up as much
ambient noise.

In order to collect data, the microphone trigger must be depressed during input.
The Visi-Pitch 6097 averages the frequencies within a chosen time frame and calculates the
frequency range. The computer interface can be adapted to analyze a maximum length of
utterance analyzable of one nrinute (Horii, 1983, p.468).

The Visi-Pitch 6087DS has four different analysis filters, depending on the voice
being measured. The filters are based on different ranges of frequencies -- Filter A: 50-
300 Hz; B: 100-600 Hz; C: 250-800 Hz; D: 500-1550 Hz -- and are selected with a switch
on the front of the instrument (Horii, 1983). The overlap in frequencies between the
different filters allows the examiner to measure different ranges that a person's voice may
fall in without excluding any frequencies. When using the Visi-Pitch to measure

fundamental frequency, determining the correct filter for the voice being analyzed is

11 Measurement of fundamental...

necessary to extract an accurate frequency measurement from the instrument. The
extracted F0 contour is displayed on the storage oscilloscope on the Visi-Pitch 6087DS. '

The Visi-Pitch 6097 and 6098 with the computer interface have identical software.
The computer interface, along with measuring fundamental frequency, is equipped with
statistical software that can calculate the frequency range, pitch perturbation, average
intensity and other parameters. The Visi-Pitch 6097 or 6098 is capable of communicating,
displaying, and manipulating the speech parameters extracted with the Visi-Pitch 608 7DS
(Visi-Pitch Manual, Kay Elemetrics, 1987). The Visi-Pitch 6087DS and 6097 interface are
used simultaneously. The computer program is menu driven, so the ranges chosen on the
6087DS must match those chosen with the computer. The frequency range, time
displayed and the display format must be selected for the 6097 computer interface prior to
data collection.

Boone and McFarlane (1988) described the Visi-Pitch as an "excellent clinical
instrument for measuring different aspects of frequency: frequency range, optimal pitch,
and habitual pitch. The Visi-Pitch offers both a digital display of frequency and an
oscilloscopic display" (p. 100). Along with measuring fundamental frequency, the Visi-
Pitch can be used in voice therapy to help facilitate new pitch levels. A patient is able to
view a wave form or a desirable pitch level on either the Visi-Pitch unit or the computer
interface and can then attempt to match the desired level.

Historv of Stroboscopy

The principle of stroboscopy was simultaneously discovered by Plateau in Brussels
and von Stampfer in Vienna in 1833. Izdebski, Ross and Klein (1990) stated that the
meaning of stroboscopy is "observing an image" (p.20). By 18 78, Oertel had extended
the use of stroboscopy to laryngology (F aure and Muller, 1992). Wendler (1992) stated
that "Oertel first observed the larynx of a living human being with the aid of a light that
was periodically interrupted by a rotating disk. By this means he was able to see the
vibration of the vocal folds" (p. 149). Izdebski et al. (1990) discuss how:

12 Measurement of fundamental...

The fundamentals of stroboscopy known since the 19th century are based on the
principle that the human retina retains an afterimage for about 200 msec.
Application of stroboscopy for laryngeal examination is therefore natural, since the
eye is not capable of discrete temporal resolution matching human voice frequency
range (p.20).

This principle of stroboscopy is derived from Talbot's Law. Wendler (1992) stated that
"According to the Talbot law, images projected on the retina leave positive after images
lasting 0.2 3. Thus, a brief sequence of individual stimuli presented at intervals <02 3
appears as a continuously moving picture" (p. 149).

The use of a laryngeal mirror to view the vocal folds has been standard practice for
many years. The combination of the laryngeal mirror and the pulsed light source has given
laryngologists the means to view both the structure and the function of the larynx. The use
of a pulsed light source such as a stroboscope to illuminate segments of the vocal folds
allows the user to see what is perceived as slow motion.

There are two different ways to utilize a stroboscope. Bless, Hirano & F eder
(1987) use the terms "synchronization" and "asynchronization" to distinguish the two types
of pulses generated by the stroboscope. These pulses can be seen when the stroboscope is
used in combination with an endoscope. When the stroboscope is used without an
endoscope, the flashes cannot be seen; but the stroboscope tracks the frequency being
produced in the same manner as it would when used in combination with an endoscope.
The frequency for the pulses is generated by the frequency being produced by the human
voice because of vocal fold vibration. Wendler (1992) described how the stroboscope

generates the light along the vibratory cycle:

Exact coincidence of the frequency of light flashes and the position of the moving
object provides a sharp and clear still picture, because points at the same phase
angle of the object's motion are repeatedly illuminated. Slight differences between
the frequency of the flashing light source and the frequency of the object's motion
produces the well-known slow motion effect (p.149).

13 Measurement of fundamental...

The first type of flashes generated by the stroboscope that can be used is the strobe
that illuminates the vocal folds at the same frequency of vibration as the frequency being
produced by the patient. Illuminated points are synchronized in time and are equal in
frequency to the "flashes" of light to provide an averaged vibratory pattern over successive

cycles. Bless et al. (1987) stated that:

Under usual circumstances, the synchronized flashes will illuminate the same
portion of the folds’ cycle for each flash with aperiodic vibration so that the folds
will appear to stand still. When movement is observable, the clinician can assume
the presence of aperiodicity in the vocal cords vibration. Thus, folds that appear to
shimmer or to continue moving under synchronized stroboscopic light indicate
aperiodic vibration ( p.91).

The aperiodic vibration is an indication that the vocal folds are not functioning normally. I
The aperiodic vibration of the vocal folds results in the perception of a harsh vocal quality,
which may be demonstrated by the frequency reading obtained by the strobe unit. One of
the characteristics typically assessed by the stroboscope is fundamental frequency (Sataloff
and Spiegel, 1988; Sataloff, 1987; Wendler, 1992). It is important that the patient be
able to produce an acoustic sound that can generate a simultaneous fi'equency in the strobe
generator. Woo, Colton, Casper and Brewer (1991) found that "A major cause of poor
quality recording was the inability of the patient’s voice to pace the stroboscope, i.e.. to
produce an acoustic source for the strobe to flash in synchrony with vocal fold vibration"
(p.234). When determining fundamental frequency based on a videostroboscopic
examination, this is a necessary factor to consider. Ifa patient has a weak voice, other
measures may be more accurate than a measure obtained from a strobe generator.

The second type of flashes generated by the stroboscope initiates theiflashes at a
slightly slower rate of vibration than the frequency produced by the patient. F aure and
Muller (1992) stated that "The problem of resolving the too-rapid motion is solved by
illuminating the larynx with brief flashes of light at a frequency just slightly less than their
vibratory rate" (p. 139). Colton and Casper (1990) explained that "At a frequency slightly

14 Measurement of fundamental...

less or greater than the frequency of vocal fold vibration (+ 2 Hz), the image is the
averaged vibratory pattern over several successive cycles and takes on the appearance of
slow motion movement of the vocal folds" (p. 180). The actual rate of vibration is not
slowed; rather the vibrations are averaged at different positions across the cycles to provide
the appearance of slow motion.

According to F aure and Muller (1992), present-day stroboscopes usually indicate
the frequency of the phonatory signal and, occasionally, its intensity. Hirano and Bless
(1986) stated that the fundamental frequency is transmitted to the stroboscope by
electronic pulses. The unit is typically equipped with either a frequency meter or a digital
display, depending on the model in use. Wendler (1992) stated that measuring the
frequency presents no problem, because it is generally done by the stroboscopic unit
(p.150). According to Wendler ( 1992), "Instruments should be made small, light, and
inexpensive. The same outlet should deliver continuous or pulsed light, as needed, and
should be adaptable for mirrors, endoscopes, and microscopes, with optional provisions for
photography and video recordings, and should provide information on the frequency and
intensity of phonation" (p. 152). The pulses are obtained via an inductive microphone that
is attached to the stroboscope and is typically held by the patient at the level of the thyroid
lamina. Izdebski et al. ( 1990) stated that "Since vibrations are pitch and intensity
dependent, evidence about voice fundamental frequency during which the image was
recorded should be provided. This can be obtained directly during the exam, or can be
extracted later on from the recorded tape and estimated by pitch matching" (p.20). The
combination of an endoscope, an inductive microphone and a stroboscope allows for the
frequency of the phonatory signal to be obtained, and both video and audio recordings are
possible.

The frequency of the strobe light is generated by the frequency being produced by
the patient. This is accomplished via the inductive microphone that is held in place against

the neck at the level of the thyroid lamina of the larynx (Colton and Casper, 1990). Baken

15 Measurement of fundamental...

(1987) described the inductive, or contact, microphone as a device that is fashioned to
perceive acoustic sigrnals, or vibrations from the body surface, and is almost totally
insensitive to airborne sounds. They are held in intimate contact with the skin by an
adhesive or strap mounting. Baken (1987) stated that the speech signal is badly degraded
by transmission through the tissues of the body, but vocal fundamental frequency is
unaltered. I
The strobe generator is able to distinguish a range of frequencies that encompass I
the typical range of human voice, allowing for use with a wide variety of patients. Wendler

( 1992) suggested that the frequency range of stroboscopes "should be sufficiently broad to

 

cover the mean speaking fundamental frequency for adults and children (SO-400 Hz)"
(p. 152). For example, the Nagashima Iaryngo-Stroboscope has a frequency range of
synchronous flashing from 50-500 Hz (Nagashima Medical Instruments Catalog).
F undamenLal EjLeguengjes of ngels

The measurement of fundamental frequency using the Visi-Pitch can be
accomplished for either steady state vowels or conversational speech. This is possible
because of the use of the acoustic microphone for data collection. The acoustic
microphone does not enter the vocal tract; therefore, it does not inhibit or change voice
productions. However, when different endoscopes are used irn combination with the
stroboscope, the voice productions that result depend on the type of scope used. Isshiki
( 1989) stated that "Vocal pitch of sustained vowel productions can easily be measured with
various instruments: stroboscope, which usually shows the vocal pitch (Hz) on a meter;
display of the sound wave on a cathode ray oscilloscope permitting calculation of the
fundamental frequency; pitch calculation on sonagram; and various types of pitch
indicator" (p.50). Conversational speech is possible with a flexible nasal endoscope, and a
patient's fundamental frequency to generate the strobe is obtained via the inductive
microphone. With a rigid endoscopic evaluation, the tongue is pulled forward resulting in

the larynx being raised in the vocal tract. This makes conversational speech impossible,

16 Measurement of fundamental...

and vowels are therefore used to measure fundamental frequency. It is important to have
some knowledge of the fundamental frequencies of vowels when measuring fundamental
frequency with both the acoustic microphone and the inductive microphone.

Vowels are produced with the vocal tract unobstructed and with the vocal folds
almost always vibrating. Vowels are classified according their articulatory dimensions:
height, frontrness, lip rounding, tongue root position and velar position (Mackey, 1987).

The English language has 13 vowel sounds. Each of ilnese vowel sounds has a particular
fundamental frequency. The tendency of high vowels (e.g., /i/ and /u/) to be produced with
higher fundamental frequency (F 0) than low vowels (e.g., /a/) in the same phonetic context
is known as the 'intrinsic pitch of vowels' (Sapir, 1989, p.44). The fundamental frequency
of vowels for men, women and children was studied in 1952 by Peterson and Barney.

This research yielded both average fundamental frequencies and the formant frequencies of
vowels. The vowel frequencies in their study were measured in the consonantal context of
I’h_d/. In the present study, the vowels /a/ and /i/ are to be used as the stimuli. Peterson
and Barney (1952) found that the average fundamental frequencies for W for men was
124 Hz and for women was 212 Hz; for the vowel til, the fundamental for men was 136
Hz and for women was 235 Hz (p.183). These data show that there is in fact inherent
differences between the vowels according to the positioning of the articulators.

The present study will determine the difference in the fundamental frequency
measurements obtained for the vowels /a/ and /i/ in isolation using the V isi-Pitch acoustic

microphone and tlnestroboscope inductive microphone.

Chapter III
Methodology

Subject Selection

A group of twenty-five persons between the ages of 24-46 were the subjects for
this study. Sixteen females, average age 31.9, rang'ng from 24 to 43 years, and nine
males, average age 33.3, ranging from 24 to 46 years, participated in this study.

Eligibility for this study was based on the following criteria as obtained through a
case history checklist:

1-no history of formal voice training for singing or public speaking
2-English is primary language

3-no history of neuropathology or cleft palate

4-non-smoker for at least the past 5 years

5-no medications that will affect voice are being taken during the study
6-no previous therapy for voice disorders

The subjects were judged with respect to overall vocal quality and pitch. Subjects were
dismissed if vocal quality was perceived by the examiner as harsh, hoarse, if any dysphonia
was present, if pitch was judged inappropriate, too high or too low for gender and age; and
if any of the criteria from the case history checklist was not met (See Appendix D). Each
subject was assigned a 1-2 digit number to insure confidentiality. Subjects were required
to sign a consent form agreeing to participate in this study after being irnformed of its
purpose (See Appendix E). Typically, the subjects that were not selected were present
smokers or those who had quit smoking within the last five years.

Data collection took place in the voice laboratory at the Otolaryngology Department
at the University of Michigan Hospital. This room is typically used for videoendoscopic
and videostroboscopic voice evaluations. The room was not sound treated, however, the
door was closed during each examination to eliminate extraneous noise that might have
been recorded by the acoustic microphone. The acoustic microphone was placed directly

in front of the subject's mouth in order to facilitate the best recording possible.

17

18 Measurement of furndarnental. ..

Data Collection

Subjects had to sustain vowel phonation for a period of eight seconds during botln
procedures. A frequency measurement was obtained for each subject using botln Visi-Pitch
6087DS and 6098 and a Bruel and Kjaer Stroboscope inductive microphones.

Both procedures involved placement of a microphone to obtain sound. The
microphones were put in place at the same time at different locations on the body in order
to obtain an estimate of the same segnent of voicing for comparison. For each vowel, fan?
and W, the subject produced four trials. The first trial was phonated for at least five
seconds to insure that the subject was capable of sustaining voicing. The trial for each
vowel was also necessary to determine whether the nnicrophones were properly placed for
data collection.

To counter balance the effects of one vowel on the other, one half of the subjects
phonated /’a/ first, the other half phonated /’i/ first. Each of the three trials that followed
were recorded and measured. The reason for using .la/ and /i/ in this study was the inherent
differences in pitch between a low back vowel, /a/, and a high front vowel, /'i/. The vowel
./a/ is typically produced with a lower fundamental frequency than the vowel /'i". The place
of production for each of these vowels is also difierent. The fundamental frequencies for
the vowels obtained with each instrument were compared. The scores obtained for the two
vowels were not compared to each other; rather, the frequency data obtained with each
instrument for the same vowel production, W or /i/, was compared.

Procedures and Instrumentation
Visi-Pitch

The Visi-Pitch 608 7DS is a self-contained analog fundamental frequency analyzer
(Bakern, 1987). Baken (1987) described how the Visi-Pitch "displays the fundamental
frequency of each cycle of the speech input, rather than the average Fo of a number of

cycles" (p.136). It was necessary to omit the onset of voicing and the termination of

19 Measurement of fundamental...

voicing using the cursors to allow for the best results possible. Each subject phonated
individual vowels, allowing the examiner to obtain a stable and constant signal of voicing.

The analysis filters for the Visi-Pitch that were used in this study were frlter A for
the men and filter C for the women. The Visi-Pitch was also capable of measuring
intensity at the same time it measures frequency. For the purposes of this study, the option
chosen was pitch only. Subjects were instructed to phonate at a comfortable speaking level
without taking a deep inhalation prior to phonation. As a method to check the calibration
of the Visi-Pitch, the examiner utilized a pitch pipe . A steady tone (middle C - 256.1 Hz)
was produced with the pitch pipe and then directed into the acoustic nnicrophone. Using
the Visi-Pitch interface statistical calculations, the measurement was calculated accurately.
The vowel productions for sixteen females were recorded by Visi-Pitch using the C range
(200-760 cps), and the productions for nine males were recorded using the A range (50-
300 cps).

Each subject was instructed to hold the Visi-Pitch acoustic rrnicrophone with the left
hand directly in front of the mouth as close to the lips as possible in order to obtain
fundamental frequency. The subject held the microphone directly in front of the mouth.

The subject was given a practice trial. The subject phonated the vowel /u/ for five
seconds or greater, to demonstrate to the subject what was required with this task. Before
the actual task began, the cursors that denote time periods were set in position. The total
time window was 8 seconds. The left and right cursors were set 1.5 seconds into the
window to frame the 5 second window from which pitch was extracted and analyzed. The
subject was required to sustain the vowels /i/ and .’a/ for as long a duration as. possible. A
frequency reading using the inductive microphone and the synchronstroboscope was
obtained simultaneously for both vowel productions.

Stroboscope with Inductive microphone
A Bruel and Kjaer model 4914 sychronstroboscope microphone was used to collect

frequency information from each subject. The synchronstroboscope flash rate tracked the

20 Measurement of fundamental...

fundamental frequency being produced by the subject (Baken, 1987) and the microphone
obtained the frequencies that were produced. The subject's F0 controlled the rate of the
flashes being produced. The subject‘s fundamental frequency was obtained through the
inductive microphone that is attached to the synchronstroboscope.

The microphone was placed on the subject's neck along the lateral aspect of the
thyroid lamina to measure the fundamental fi'equency of the voice (Colton and Casper,
1990). Each subject held the inductive microphone (Bruel and Kjaer
synchronstroboscope, model 4914) with the right hand against the neck at the level of the
thyroid lamina. The choice of which side to place the inductive microphone was dictated
by the positioning of the instruments in the room.

Prior to any data collection, each subject was instructed to produce a steady state
vowel in order to determine whether the inductive microphone was properly placed. The
subject produced the vowel /'a/ for approximately 8 seconds while holding the inductive
microphone in place. The Visi-Pitch acoustic microphone was sanitized with an alcohol
pad prior to use by each subject. The examiner explained to each subject that the vowels
needed to be phonated for at least eight seconds each. Each subject was able to see the
computer screen as it was set up for data collection. Each subject was instructed to
"Phonate the vowels /a/ and /i/ in both a comfortable speaking pitch and at a comfortable
loudness level".

With the inductive microphone in position, the subject phonated /i/ and /a/,
sustaining duration for approximately eight seconds. The subject produced three trials for
each vowel while the same recording was obtained with Visi-Pitch. I
Data Ana sis

Each subject produced six individual phonations: three phonations of /a/, and three
phonations of /'i/'. A comparison of the scores was obtained and were calculated for each

production for each instrument

21 Measurement of fundamental...

Statistical Ana_lysis:

A paired t-test was used for statistical analysis. The t-test is used for measured
variables when comparing two means. The paired t-test compares two paired observations
on the same individual or on matched individuals. In this case, productions of /a/ on the
Visi-Pitch microphone were compared with productions of la/ obtained with the inductive
microphone; the same was true for lil. The averages for /a/ and /i/ were determined using
measures of central tendency.

A Pearson correlation was calculated to determine the strength of the relationship
between the two variables. This type of analysis detemnined the linear relationship between
the two variables.

Measures of central tendency and dispersion of scores were also analyzed.
Measures of central tendency provided average frequency scores for the sample for each

microphone score.

Chapter IV
Results

M

The present study was desigred to determine whether the measurement of
fundamental frequency for the vowels /a/ and /i/ with the acoustic microphone and the
inductive microphone are different. It was important to determine if the differences
between the measurements obtained with these microphones are relevant when measuring
fundamental frequency in clinical practice.

Average fundamental frequency data for the vowels /a/ and .I'i/ were calculated for
the 9 males and 16 females, N: 25, as measured with the Visi-Pitch acoustic rrnicrophone
and the inductive microphone attached to the B& K synchronstroboscope. Results are

summarized in Table 1 and displayed in Figures 1a and lb.

23 Measurement of furndamental...

Table 1. Mean fundamental frequency measurements for males, females, and the sample

population for each vowel production using the Visi-Pitch acoustic microphone (Ac) and.

 

 

 

 

 

the Inductive microphone (1nd).

Stimuli Ma es E£m_al§ SI; fl otal Avemge (m)
Ac /a/ Trial 1 114.03 208.9 57.88 174.74
Ac /a/ Trial2 113.31 209.1 57.34 174.62
Ac /a/ jlrjal3 115.62 210.5 57.88 176.35
Ind. /a/ Trial 1 115.16 209.38 57.63 175.46
Ind. fax' Trial 2 113.91 210.01 56.24 175.42
Ind. /a/ Irial3 116.41 211.29 57.86 177.14
Mgrms [a/fi 114.74 209.86 57.47 175g;
Ac /i/ Triall 122.75 218.31 57.61 183.91
Ac /i/ Trial 2 123.41 219.2 57.57 184.72
Ac .Ii/ Trial 3 123.58 222.025 59.77 186.59
Ind. x’i/ Trial 1 122.53 218.95 58.06 184.24
Ind. li/ Trial2123.67 220.29 58.26 185.51
Ind. /i/' Tnfl3124.45 222.35 59.72 187.11
Means li/ 123.39 220.19 58.50 185.35

24 Measurement of fundamental...

As demonstrated in Table l, the productions for each vowel over the three trials are
consistent. Intrasubject variability between the average frequency productions is typically
within 5 cycles per second. The mean frequencies shown in Table 1 for each vowel an
each rrnicrophone demonstrate minimal intersubject variability between productions. In
addition, most subjects demonstrated a pattern of the 2nd and 3rd productions for both
vowels and for both microphones either both being higher or both being lower in
frequency than the first (See Appendix A). This pattern is reflected in Table 1 for the
females for both vowels, and for the males for the vowel /i/. Figures la and 1b show the
mean frequencies for males and females. These figures demonstrate the frequency

differences between males and females for the vowels /a/ and /i/.

Figure 1a. Mean frequencies for males and females and the sample population for the

phoneme /a’.
Mean F0 for /a/
2507
I.
2001
H 3
e 150-;
r I
t IOOI if;
50+. 157,.

    

          

}' j' ..

Ind’a/3

{-7.
i

Ind/a2

. 31'? ‘* gr I

Acfa/l Ada/2 Ac/av’3
Trials

- Males B Female 5:] Total

 

In’a/l

25 Measurement of fundamental...

Figure 1b. Mean F0 for males and females and the sample for the phoneme /'i>".

Mean F0 for /i/

250~I
200%

H I
e 1504'
r .
t 100i
2 ‘1
50f

     

Ac/‘l Ac r2 Adi/3 Ind/ill Ind/i/Z Ind/U3
Trials
- Males Female [:3 Total

It was necessary to determine the average fundamental frequencies for each subject
for each vowel in order to determine the differences between the frequencies as measured
with each microphone. Raw data for each subject's fundamental frequencies as measured
with each microphone are indicated Appendix A. The fundamental frequency differences
between the measurements obtained with each microphone were calculated for each of the
subject's six vowel productions. The mean differences of the difference (MDD) was
calculated for each subjects three trials of /’a/ and three trials of /if. The average diflerences
between the fundamental frequency measurements by the acoustic microphone and the

inductive microphone are shown in Table 2.

Measurement of fundamental...

Table 2. Mean Differences of the calculated differences for 3 trials for each subject.

 

S .. 1g" 3.4. ii"
1 0.73 -0.8 13 0.76 1.3
2 -0. 16 0.83 14 0.80 0.46
3 0.56 0.56 15 0.80 1.125
4 0.75 0.9 16 -0.70 1.03
5 0.93 0.73 17 0.83 -3.48
6 0.92 0.9 18 0.63 0.59
7 0.43 -l.86 19 1.23 1.46
8 0.56 0.23 20 0.86 2.26
0.83 0.86 21 0.77 0.44
10 1.9 1.06 22 0.36 1.96
11 0.43 0.575 23 1.46 0.26
12 0.8 -0.53 24 0.97 0.30
25 0.67 -0.09

27 Measurement of fundamental...

The mean of the frequency differences (MFD) and the standard deviation were
determined in order to perform the Paired t-test. The Paired t-test was used to examine

these differences by using the following formula:

t = X f SD/ \IN

 

The differences are between the frequency measurements obtained for either fat" or .I’i/
using the acoustic and inductive microphones. As demonstrated in Table 2, the differences
are consistently small, within 1-2 Hz for both vowel productions for all subjects. The
greatest difference noted is -3.48 for the vowel /i/' for subject 17. The t-score is high as a
result of the small differences. The small differences indicate that the data have a narrow
distribution and therefore result in a small SD. The MFD is then divided by that small SD
for that vowel. Because the MFD is a value greater than the SD, the resulting value is
high. Ifthe distribution had been wider, the resulting value would have been a low t-score
because the mean and the SD would have been more similar in value. The t-value results

are summarized in Table 3.

Table 3. t-values for /a/ and .I'i/ for entire sample (N: 25).

Measures W /i/
Mean of the Differences .7248 .4428
Standard Deviation .481 5 1. l 727

t-values 7.526 1.887

28 Measurement of fundamental...

Figure 2. MFD between microphones for all subjects.

Mean Fo Differences between Microphones

 

 

3* \. Ag
D \ l *- g \. A t I r. ¥ X
I 0W i l I \\ I I w.
f 4 n 4’ 4" 4‘ I
1" ............................... / ............................................. t ..... j .....................................
f I \ I i
i \i n‘ (i
.24! ............................ If? ............................................... 1. . .r ......................................
; l i
I I I
.31 .............................................................................. LIL .......................................

n" *2'“ 34 '5' "s “r ”8” 9'40'1ii‘2' n'an'rn‘s'n's'nv"19"1420"21"‘2‘2"23“2'4“2s
Subjects
‘0 113! + A!

Based on a two-tailed t-test, assuming that the alpha error was occurring at both ends of
the distribution, and using alpha level of .01, any value greater than 2.797 (Linton and
Gallo. Jr., 1975) for 25 subjects would be statistically significant. Alpha level is the
probability of rejecting the null hypothesis (Linton and Gallo, Jr., 1975). The tabled value
2.797 is derived fiom the t-value relative to the degrees of freedom (df =24). The t-value
for Kai" for the fundamental frequencies obtained with the acoustic microphone and the
inductive microphone is 7.526, which is greater than 2.797. This t-value indicates that
there is a statistically significant difference in the average fundamental frequencies for W as
measured with the two instruments Via the microphones and their input. However, the t-
value of 1.887 for /i.’ is less than 2.797, indicating that the differences are not statistically
significant for this vowel. It should be noted these results change when the results from
subject #17 are eliminated from the Paired t-test. This change illustrates that the

differences in fundamental frequency measurement of these two vowels by the acoustic

29 Measurement of fundamental...

microphone and the inductive microphone are both statistically significant at the .01 alpha
level. The difference between frequency measurements for this subject for the vowel ./i/ on
the first trial is -8.1 (See Appendix A). This number creates a bias that is reflected in a low

t-value. The t-values for an N = 24 is demonstrated in Table 4.

Table 4. t-values for sample size N=24.

Measures r’ar' /i/
Mean .7204 .6063
St. D .4913 .8591
t-values 7. 175 3. 45

With these t-values calculated. the degrees of freedom in this case would be 23, and the
critical value therefore changes to 2.807. The resulting t-values when #17 is excluded
from the calculations demonstrates a statistically significant difference that allows the
examiner to reject the null hypothesis stating that there is no difference in the fundamental
frequency measurements as obtained via the acoustic and inductive microphones.

The difference across instruments for each subsequent trial for each vowel is less
than or equal to a 1.0 Hz for 67% of the total trials; 33% were greater than or equal to a
1.0 Hz difference, with the geatest difference beirng noted for subject #17's first trial of .’i/ ,
which is 8.1 Hz . The averages of the frequencies produced for both vowels with both
microphones were calculated for each subject. Results are shown in Appendix B. Average

fundamental frequencies for subjects 11, (F 0 /a/ = 86.22 and x’i/ 83.62), and 20 (F0 W =

348.27 and fi’ =343.34), were the lowest and the highest, respectively.

30 Measurement of fundamental...

The range of the frequencies produced by all of the subjects was calculated. Table
5 demonstrates the ranges calculated for each of the vowel productions as measured with

botln microphones.

Table 5. Ranges for the mean F0 produced.

Acoustic R (R - outlier) Inductive R (R - outlier)
.x‘a/l 262.7 168.6 /a/1 262.9 168.6
lai’2 259.3 154.4 ."a/2 260.4 155.7
/a/3 263.5 154.9 /a/3 263.5 155.5
/i/1 263.4 172.4 /i/1 264.3 173.7
.z’i52 254.4 180.1 x’i/2 258.5 180.1
I'i/3 258.8 181.1 I'i/3 258.9 180.5

Subject 20 produced fundamental frequency averages well above the other subjects
averages. For each production, the range was calculated including the outlier (subject 20)
and excluding the outlier. The range (represented by R), which determines the width of
the entire distribution, was affected by this subject's productions. The range value without
the scores from subject 20 was not as wide. The range values for each production without
subject 20 indicated a more narrow distribution of scores than the R for the entire sample.
Results of the measures of central tendency are demonstrated in Table 6. Measures of
central tendency were calculated using the entire sample of 25. The measures of central
tendency show where the data were clustering. The mean fundamental frequencies for the

total sample for botln /'a/ and /i/ were

31 Measurement of fundamental...

Table 6. Measures of Central Tendency for entire sample (N = 25).

___utsMcas _A_c_/al Ind/at 5931 my
X 175.2 176.0 185.0 185.62
SD 56.71 57.24 58.31 58.68
R 267.1 266.3 262.6 264.3

The mean value for the entire sample for both vowels shows that the frequencies
measured with the two difierent microphones are within 1 Hz of each other. The standard
dexiation showed how widely the frequencies produced varied from the mean. The range
showed the difference between the highest frequency and the lowest for each vowel with
each microphone.

A Pearson correlation was also calculated to determine the strength of the
relationship between the average of the 3 frequencies measured using the Visi-Pitch
acoustic microphone and the inductive microphone with the synchronstroboscope. The
correlation coefficient for the vowel /a.’ is r = +.99991. The coefficient for the vowel fit’ is
r = +.9998. These numbers indicate an almost perfect positive correlation between the two
sets of measurements. According to Sprinthall (1990). correlations can provide better than
chance predictions. If two events are correlated, then a knowledge of one of those events
allows a researcher to predict the occurrence of the other (Sprinthall, 1990). In this case,
the events, or the different microphones, are closely related in the frequency information
that is recorded. Clinically, these two microphones extract frequency information that
would complement each other, not contradict.

The fundamental frequencies produced by the majority of the subjects in this study
fall within ranges of fundamental frequency norms as found in previous studies. Although
there is no hard and fast rule for what an individual's fundamental fi'equency should be,

ranges are typically used to determine whether one compares favorably with other persons

32 Measurement of fundamental...

of the same age and gender. In the present study, the stimuli were steady state vowels.
Fundamental frequency is frequently discussed in terms of speaking fundamental frequency
(SFF). Many of the norms that are in place have been determined via measurement of
frequency when a patient is reading a passage, such as the "Rainbow Passage" (Fairbanks,
1937). However, the results from the present study can be generalized to fit the SFF
norms (See Appendix C). As with the males subjects from the Hollien and Shipp ( 1972)
study, the averages from Stoicheffs study were determined according to speaking
fundamental frequency. According to the norms from Hollien and Shipp (1972) and
Stoicheff ( 1981), the average fundamental frequencies for both male and female subjects

inn the present study are within normal limits for their age ranges and gender.

Chapter V

Discussion, Summary and Conclusions
mm

In the clinical evaluation of voice, obtaining a measure of a patient's fundamental
frequency is often necessary. The instruments that are used to measure fundamental
fi'equency vary from using a pitch pipe as a pitch matching device to using instruments
such as a spectrograph to view the wave form to make specific determinations and
calculations about fiequency. As technology changes, it is important to determine whether
or not the instruments are measuring fundamental frequency accurately. In other words,
are the instruments used in clinical practice valid?

The goal of the present study was to determine whether the fundamental frequency
measurements obtained with the Visi-Pitch acoustic microphone and the
Synchronstroboscope inductive microphone differ fi‘om each other in a systematic way.

The purpose of the present study was to determine whether any differences were
present between the measurements obtained from two different microphones and not to
identify any specific pathology. The stimuli for this experiment were the vowels /a/ and /i/,
to be phonated by each subject three times each. To counter balance any effects one
vowel rrnight have on the other, 12 of the subjects phonated /a/ first, 13 phonated /i/ first.
The number of subjects for the study totalled 25, comprised of 16 females and 9 males.
For each subjects six vowel productions, the inductive microphone was held with the right
hand against the neck at the level of the thyroid lamina, and the acoustic microphone was
held with the left hand directly in front of the mouth as close to the lips as possible. The
microphones were utilized simultaneously to facilitate measurement of the same segnent of

voicing.

33

34 Measurement of fundamental...

Wags

Fundamental frequency of steady state vowels was measured with two instruments:
the Visi-Pitch via the acoustic microphone and the synchronstroboscope via the inductive
rrnicrophone. The two microphones, although both measure fundamental fi'equency, are
different mechanically. The acoustic microphone measures fundamental frequency
through the sound pressure levels produced by the subject. The inductive microphone
measures fundamental frequency through the vibrations produced iii the larynx during
phonation. The inductive microphone, unlike the acoustic microphone, does not measure
sound pressure levels; extraneous noises, therefore, do not effect the fundamental
frequency.

In the present study, the fundamental frequency measurements of the steady state
vowels /a/ and /i/ were found to have statistically significant differences between the
frequencies for the two microphones. A pattern emerged whereas 94% of the
measurements obtained with the inductive microphone were slightly higher than the
measurements with the acoustic microphone. When the differences for all 25 subjects
were averaged, it was found that the differences for /a/ were statistically significant, but the
differences for /'i/ were not. This discrepancy was the result of the data from one subject.
With that subject's data removed from the total, it was determined that the differences for
both /'a/ and /i/ were, in fact, statistically significant. However, although the fundamental
fi'equencies were higher, most were less than or equal to 1.0 cps higher. Rejection of the
null hypothesis according to the paired t-test would suggest that there is a statistically
significant difference in the measurement of fundamental frequency as measured with the
acoustic microphone and the synchronstroboscope inductive microphone.

Although the findings from the present study were statistically significant,
differences of approximately 1-2 Hz between microphones would not be considered
clinically sigrificant during a voice evaluation. Considering the minimal differences in the

acoustic microphone , the fundamental frequency measurement with the inductive

 

35 Measurement of fundamental...

microphone is a valid measure of fundamental frequency as it measures what it purports to
measure.

In clinical practice, fundamental frequency is measured for both steady state
vowels and conversational speech. The acoustic microphone is capable of measuring
fundamental frequency for botln types of stimuli. The Visi-Pitch is described in the
literature as an accepted instrument for the measurement of different aspects of frequency
as well as a clinical therapy tool (Boone & McFarlane, 1988; Horii, 1983). This is because
the acoustic microphone in no way impedes the vocal tract. On the contrary, the inductive
microphone is used clinically with videoendoscopy or videostroboscopy. Much of the
literature states that this frequency reading from the synchronstroboscope is the patient’s
fundamental frequency. In various studies utilizing the synchronstroboscope with the
inductive microphone, it is stated that the fundamental frequency measurement is not a
problem to extract because it done by the stroboscope (F aure & Muller, 1992; Izdebski,
Ross and Klein, 1990). Wendler (1992) specifically stated that "Measuring frequency
presents no problem, because it is generally done by the stroboscopic instrument" (p. 150).
The present study utilized steady state vowels as stimuli, as is often used in endoscopic
examinations. Quite often, when a flexible nasopharyngoscope is utilized for vocal fold
visualization, it is possible to have the patient phonate vowels in order to examine the
vibration during conversational speech because the flexible endoscope does not interfere
with the normal functioning of the vocal tract. However, when a rigid endoscope is
utilized, the shape of the vocal tract is changed and the larynx is elevated due to the tongue
being pulled forward during examination. This obviously makes speech impossible. The
resulting fundamental frequency as measured with the inductive microphone may be
inaccurate because of the changed shape of the vocal tract. The user of these instruments
must be cognizant of the stimuli being used to elicit fundamental frequency. In the context
of steady state vowels, fundamental frequency can be measured accurately with these

instruments.

36 Measurement of fundamental...

The findings of the Pearson correlation would concur that there is a strong, positive
relationship between the microphones for the measurement of steady state vowels. The
correlation for the vowel /a/ is r = +.99991, and for the vowel /'i/ is r = +9998. These
coefficients indicate an almost perfect positive correlation between the frequency
measurements with the acoustic microphone and the measurements with the inductive
microphone. These correlations would indicate that with the measurements from either the
inductive or acoustic microphone, it is possible to predict the results that the other will
produce.

The mean fundamental frequency for the males, n=9, for all trials for /a/ and /i/,
was equal to 119.07 Hz. According to Hollien and Shipp (1972), this average is within
normal limits for speakirng fundamental frequency for males. The females in the present
study, n=16, produced a mean fundamental frequency of 215.06 Hz. Stoicheff (1981), in
a study of SFF, found results similar to that of the present study. Altlnough the present
study utilized steady state vowel phonation, it can be generalized that the subjects were
producing these vowels in a habitual pitch, much the same as the subjects produced
speaking fundamental frequency in the previous studies. The findings of the present study
are irn ageement with previous studies of the fundamental frequencies of vowels (Peterson
& Barney, 1952; Sapir, 1989). It was found that the fundamental frequency productions
for the vowel /i/ were consistently higher than for the vowel /a/ for botln males and females.
In previous studies pertainirng to the measurement of the fundamental frequencies of

vowels, this was also found to be the case (Peterson & Barney, 1952; Sapir, 1989).

Conclusiggs
Based on the results of the present study, it may be concluded that although there is
a statistically significant difference between the fundamental frequencies as measured with

the two microphones, although the difference is not considered to be clinically significant.

37 Measurement of fundamental...

In the measurement of furndamental frequency, a difference of < 1 cps does not afiect the
results of a clinical evaluation in an adverse way. Clinical application, unlike speech
science and basic research, does not require frequency specificity of this degee.
Conversational speech is a range of frequencies; thus the cyclic limits are not clinically
practical. Further, the irnstruments were shown to be highly correlated, indicating that the
frequency measurements via the acoustic and inductive microphones are strongly related.
Future research might focus on whether fundamental frequency measurements differ when
using flexible nasopharyngolaryngoscopy and rigid endoscopy to determirne the validity of
the frequency measurement with the inductive microphone when the vocal tract is

altered.

38

GLOSSARY

1. amplitude - the amount of vibratory displacement; how far an object moves describes its
amplitude. (Y ost and Nielson, 1985 p. 7).

2. Cycles per second - the frequency of a sinusoid is the number of cycles a sine wave
completes per unit of time, for instance, the number of times per second an object moves
back and forth. (Y ost and Nielson, 198 5, p.4) A cycle is the complete sequence of values
of a periodic quantity that occurs during one period.

3. decibel - a quantitative unit of relative sound irntensity or sound pressure. Referent =
.0002 dynes/cm2 (Zemlin,1988)

4. direct laryngoscopy - an invasive surgical procedure requiring that the person be
anesthetized. Direct laryngoscopy permits more detailed examination of laryngeal
structures, including the actual manipulation (Colton and Casper, 1990, p.177).

5. endoscope - an instrument inserted intraorally or intranasally to view the vocal folds.

6. fiberoptics - a thin transparent homogeneous fiber of glass or plastic that is enclosed by
material of lower index of refraction and transmits light throughout its length by internal
reflections; a bundle of such fibers used in an instrument for bending light or seeing around
comers (Webster, 1973, p.424).

7. input filtering -tlnere are three types of filters: low pass- which attenuates high
fi'equency energy but passes low; high pass- which attenuates low frequency energy but
passes high; and octave band - which attenuate a band of high or low frequencies, passing
the remaining. I

8. Flexible Nasopharyngolaryngoscopy- the use of a flexible fiberoptic endoscope used
for viewing the vocal folds.

9. frequency - the number of complete vibrations or cycles per unit time and is usually

measured in Hz or cps.

39

10. fundamental fiequency (F o) - fundamental frequency refers to the most frequently or
commonly occurring frequency that characterizes a particular vocal production (Colton
and Casper, 1990, p. 184). The fundamental is the lowest component of a complex tone.
11. harmonics - are whole number multiples of the fundamental frequency. These tones,
which occur over the fundamental, are called harmonics 0r overtones.

12. Hertz (Hz) - most commonly used symbol to denote frequency.

13. indirect laryngoscopy - the traditional means of examining the larynx through the use
of a laryngeal mirror (Colton and Casper, 1990, p.176)

14. intensity - Intensity is the physical property of an acoustic signal, and the loudness of
that signal is related to its intensity (Borden and Harris, 1984).

15. invasive - an instrument, such as an endoscope, is inserted into a body cavity.

16. larynx - The larynx is a tube with several folds. It is composed of ligaments, muscles,
and mucous membranes. Its framework is cartilage, and it is suspended in the neck by a
series of flat strap muscles. Its superior support is via the nonarticulated hyoid bone, and it
is connected inferiorly to the trachea, just superior to the first tracheal ring. Nervous
supply to the larynx is provided by vagus, the tenth cranial nerve; and its blood supply is
from a branch of the common carotid artery (Stemple, 1984, p.13).

17. light source - in conjunction with the use of an endoscope, a steady stream of light or a
pulsed light (strobe). The light is generated externally to the endoscope, and travels
through the endoscope via fiberoptic cables.

18. loudness - Loudness is the subjective impression of the power of sound (Martin, 1986,
p.55). I

19. mucosal wave - a wave-like motion of the vocal folds is seen on the superior surface
(lamina propria) in normal speakers when examined with an endoscope. This wave-like
motion reflects their complex structure and movement. Any change irn tine normal
movement patterns will be reflected in the appearance of the mucosal wave

(Colton and Casper, 1990, p.28). Mucosal wave is a stroboscopic phenomenon.

40

20. non-invasive - a measurement procedure that does not involve placing any instrument
into a body cavity. Sound is measured acoustically.

21. nasopharynx - the area where the back of the nose and the throat meet, at the level of
the soft palate. Bounded above by the rostrum of the sphenoid bone and the pharyngeal
protuberance of the occipital bone, the nasopharynx is limited inferiorly at the level of the
soft palate. It communicates anteriorly with the posterior nares or the choanea of the nasal
cavities and laterally with the pharyngeal orifice of the auditory tube (Zennlin, 1988,
p.223).

22. phase - a term used in describing vibratory as well as wave motion and is useful in
describing the relationship between two or more vibrations or wave motions. Phase may
be defined as the portion of a cycle through which a vibrating body has passed up to a
given instant (Zemlirn, 1988, p.412). Phase symmetry of the vocal folds refers to
differences in appearance between the moving vocal folds (Colton and Casper, 1990, p.
29).

23. pitch - Pitch is the psychological subjective perception of frequency.

24. quality - The sharpness of resonance of a sound system. The vividness or identifying
characteristics of a sound (Martin, 1986, p.55). The psychological perception of a wave
form.

25. Rigid Videostroboscopy - a procedure for viewing the vocal folds at a perceived
slower rate, and capable of emulating fundamental frequency.

26. stroboscopy - the use of a strobe light source in combination with an endoscope. The
vibratory image portrayed by stroboscopy appears to be in slow motion but is, instead, a
composite image pieced together fi'om many different cycles (Watterson and McFarlane,
1991, p. 81 ).

27. Visi-Pitch - a portable instrument that extracts and displays frequency in real time,

either from a live voice via a microphone or from a tape recording. (Horii, 1983)

41

28. vocal fold - Each fold consists of a bundle of muscle tissue; The main part of the vocal
cord consists of the vocalis muscle. The vocalis muscle is covered with the elastic conus, I
which is markedly thick near the edge of the fold, comprising the vocal ligament.
Superficial to the elastic conus lies the mucous membrane (mucosa) which consists of the
very thin epithelial layer and the lamina propria (Hirano, 1974, p. 1-3). The vocal folds
are located irn the larynx and appear shelflike when viewed from a superior angle. The
vocal folds vibrate when air from the lungs is passed between them, in the area called the
glottis, in order to produce sound. The vocal folds can also act as a protective valve for the
trachea (airway), closing off to protect against foreign objects (such as food or liquid
during a swallow).

29. vocal tract - an extended tube starting immediately within the larynx and extending up
the throat into the oral cavity, coupled with the nasal cavity. The resonating tubes of the
vocal tract include the hypopharyrnx, the oropharynx, and the nasal cavities. Sound is
produced at the level of the vocal folds, and vibrates within these resonating cavities.

(Boone and McFarlane , 1988).

APPENDICES

APPENDIX A

42

APPENDIX A

Averages for each subject's vowel productions: Ss 1-12

 

 

 

 

 

 

 

 

 

 

 

 

_S_ti LE VP /a/ Inductive /a/ w, VP /i/ Inductive ."i/ Diff.
S 1 1 173.8 175.0 +1.2 180.7 179.5 -1.2
l 2 187.1 188.0 +.90 178.9 179.3 +.40
1 3 174.7 175.7 +.10 195.5 193.87 -l.625
S 2 1 178.9 179.4 +.50 183.1 184.1 +1.0
2 2 177.1 177.75 +.40 199.1 198.5 +60
2 3 177.4 176.0 -1.4 189.5 190.4 +.90
S 3 1 182.6 183.1 +.50 193.2 193.5 +30
3 2 187.3 188.0 +.70 190.8 192.1 +1.3
3 3 188.0 188.5 +.5 192.9 193.0 +.10
S 4 1 185.8 186.0 +.2 192.0 192.8 +.80
4 2 187.7 188.75 +1.05 193.1 194.6 +1.5
4 3 190.5 191.5 +1.0 193.9 194.3 +.40
S 5 1 253.4 253.8 +.40 253.8 255.2 +1.4
5 2 240.5 241.8 +1.3 263.5 264.2 +.70
5 3 242.4 243.5 +1.1 260.1 260.4 +.10
S 6 1 193.4 194.75 +1.35 197.1 198.5 +1.4
6 2 192.2 192.8 +.60 195.4 196.3 +90
6 3 192.5 193.3 +.80 195.2 195.6 +.40
S 7 1 205.4 206.0 +.60 230.4 230.8 +.40
7 2 215.4 215.6 +.20 230.3 231.3 +1.0
7 3 218.9 219.4 +.50 228.3 2222 -6.1
S 8 1 220.5 221.2 +.70 211.8 211.5 -.30
8 2 221.9 222.2 +.30 215.6 216.8 +1.2
8 3 218.1 218.8 +.70 214.8 214.6 -.20
S 9 1 101.4 103.0 +1.6 108.7 109.0 +30
9 2 98.4 98.8 +.40 106.8 107.6 +.80
9 3 97.0 97.5 +.50 106.1 107.6 +1.5
S 10 1 112.8 117.4 +4.6 133.1 134.3 +1.2
10 2 117.2 117.8 +.60 135.7 136.1 +.40
10 3 L188 119.3 +.50 137.6 139.2 +1.6
S 11 1 84.4 85.2 +.80 81.4 81.5 +.10
11 2 86.1 86.1 = 83.4 84.125 +.725
11 3 87.5 88.0 +.50 85.2 86.1 +.90
S 12 1 121.1 122.1 +1.0 143.4 143.7 +.30
12 2 125.3 125.8 +.50 142.0 139.6 -2.4
12 3 127.2 128.1 +.90 143.3 143.8 +.50

 

43

 

 

 

 

 

 

 

 

 

 

 

 

Appendix A (cont'd)
Ss13-25
S 13 1 150.0 150.1 +.10 152.6 154.3 +1.7
13 2 142.2 143.3 +1.1 154.6 155.6 -1.0
13 3 141.3 142.4 +1.1 151.4 154.6 +3.2
S 14 1 119.2 120.1 +.90 131.7 132.5 +.80
14 2 120.8 121.7 +.90 142.0 142.4 +.40
14 3 134.0 134.6 +.60 140.6 140.8 +.20
S 15 1 99.7 100.2 +.50 107.1 107.75 +65
15 2 100.4 101.5 +1.2 103.9 105.625 +1.725
15 3 100.3 101.0 +.70 102.1 102.0 HQ
S 16 1 120.8 121.0 +.20 119.1 120.2 +1.1
16 2 114.6 114.8 +.20 123.4 124.5 +1.1
16 3 117.9 115.4 -2.5 125.8 126.7 +.90
S 17 1 116.9 117.4 +.50 127.7 119.6 -8.1
17 2 114.8 115.4 +.60 118.9 117.5 -1.4
17 3 116.6 118.0 +1.4 120.2 119.25 -.95
S 18 1 206.9 207.4 +.50 242.9 243.375 +.475
18 2 211.3 212.125 +.825 244.7 245.7 +1.0
18 3 214.8 215.375 +.575 266.3 266.6 +.30
S 19 1 206.2 206.5 +.30 210.8 211.85 +1.05
19 2 205.6 206.7 +1.1 209.1 210.83 +1.73
19 3 206.2 208.5 +2.3 207.6 208.0 +1.6
8 20 1 347.7 348.1 +1.0 344.8 345.8 +1.0
20 2 345.4 346.5 +1.1 337.8 342.6 +4.8
20 3 351.0 351.5 +.50 344.0 345.0 +1.0
S 21 1 199.0 199.83 +.83 200.9 201.83 +93
21 2 196.5 197.1 +.60 202.0 202.2 +.20
21 3 196.4 197.3 +.90 205.0 205.2 +220
S 22 1 200.3 199.57 -.73 213.7 213.6 -.10
22 2 206.9 207.6 +.70 225.4 230.2 +4.8
22 3 207.9 209.0 +1.1 234.2 235.4 +1.2
S 23 1 177.8 178.1 +.30 197.0 197.5 +.50
23 2 177.5 179.0 +1.5 183.8 184.1 +30
23 3 176.2 178.8 +2.6 168.6 168.6 =
S 24 1 214.0 214.83 +.83 209.8 210.5 +.70
24 2 191.2 192.4 +1.2 213.5 213.6 +.10
24 3 199.7 200.6 +.90 230.5 230.6 +.10
S 25 1 206.0 206.3 +.30 222.4 223.16 +.76
25 2 208.4 209.6 +1.2 218.0 217.0 -1.0
25 3 216.7 217.2 +.50 222.9 222.86 -.04

 

APPENDIX B

1‘»)

LII-F0)

\OOOHON

APPENDIX B

44

Mean F o for subjects three vowel trials and mean frequency differences.

/a/ Ac
178.5
177.8
185.9
188.0
245.4
192.7
213.2
220.1
98.93
116.27
86

la/ Ind
179.5
177.6
186.5
188.75
246.3
193.6
213.6
220.73
99.77
118.17
86.43
125.33
142.27
125.47
100.9
117.07
116.93
211.63
207.23
348.7
198.08
205.39

Diff.
+1.0
-.20

+60
+.7S
+90
+90
+.40
+.63
+.93

+.77
-.60
+83
+63
+1.23
+.87
+.78
+.36

/i/ Ac
185.0
190.5
192.3
193
259.1
195.9
229.6
214.06
107.2
135.47
83.33
142.9
152.87
138.1
104.37
122.77
122.27
251.3
209.1
342.2
202.3
224.4

/i/ Ind
184.93
191
192.86
193.9
259.9
196.8
228.1
214.3
108.07
136.53
83.91
142.37
154.83
138.57
105.12
123.8
118.78
251.89
210.23
344.47
203.08
226.4

Diff.
-.07

+1.96
+.47
+.75
+1.03
3.49
+59
+1.13
+2.27
+.78
+2.0

177.17
201.63
210.37

45

Appendix B (cont'd)
178.63 +1.46 183.1 183.4
202.61 +.98 217.9 218.23
211.03 +.66 221.1 221.01

+.30
+33
-.09

APPENDIX C

46

APPENDIX C

Norms for Speaking Fundamental Frequency according to age ranges and gender

Hollien and Shipp (1972) Norms for Males:

AGES Norms for Males
20-29 F 0 =120 Hz
30-39 F 0 =112 Hz
40-49 F 0 =107 Hz
Present study finding for males for sustained vowel phonation:
Ages /a/ /i/
20-29 126 Hz 144 Hz
30-39 117 Hz 129 Hz
40-49 101 Hz 110 Hz

In a study of SFF characteristics of nonsmoking female adults, Stoicheff (1981) study of
nonsmoking female adults found:

Ages Norms for Females
20-29 224.3
30-39 213.3
40-49 220.8.

Present study findings for females for sustained vowel phonation:

Ages /a/ W
20-29 210 Hz 224 Hz
30-39 211 Hz 218 Hz

40-49 206 H2 218 Hz.

APPENDIX D

to

4;.

kl!

O\

47

APPENDLX D

Case History Checklist

Name
Subject Number

Age
Gender M F

. Have you ever had any formal voice training or public speaking training?
yes no

. Is English your primary language?
yes no

. Do you have any history of neuropathology 0r cleft palate?
yes no

. Have you been a non-smoker for at least the past 5 years?
yes no

. Are you presently taking any medications that will or do affect your voice?
yes no

. Have you had any previous therapy for any voice disorders?

yes no

APPENDIX E

48

APPENDIX E

Informed Consent

I agree to participate in a study of the reproduction of human voice and for the analysis of
a selected acoustic parameter, specifically fundamental frequency, associated with normal
voice production. I understand that I will be asked to produce a series of vowels and that
recordings of my voice will be obtained for analysis via an acoustic microphone that will be
hand held, and an inductive microphone that will be placed on the side of my throat and
held in place with a velcro strap. Participation in the study is expected to last
approximately 20 minutes.

My signature below indicates that I have read this consent form and have been advised
about the following:

--I understand that my participation in this study is strictly voluntary;

--I understand that there will be no financial compensation for my
participation in this study;

--I understand that there are no fees or charges for any of the procedures;

--I understand that I can withdraw voluntarily from the study at any time
without penalty;

--I understand that the results of my personal performance irn this study will
be kept confidential;

--I have been advised that it is likely that the results of the study will be presented in
an educational and/or professional setting and that my name will remain

confidential at all times.

I have read this consent form and have agreed to participate in the study.

(Signature)

(Date)

REFERENCES

49

REFERENCES

Baken, R. J. (1987 ). Vocal Fundamental Frequency. Clinical Measurement of Speech and
Voice. Boston: College Hill Press.

Bless, Diane M., Hirano, M., & Feder, Robert J. (1987 ). Videostroboscopic Evaluation of
the Larynx. Ear Nose d oat Jo

Boone, Daniel R. & McFarlane, Stephen C. (1988). 2111c Vpice epd Voiee Therapy.
Englewood Cliffs, NJ: Prentice Hall.

Coleman, Robert F. & Markham, Ira W. (1991) Normal Variations in Habitual Pitch.
10mg of Voice, 5, 173-177.

Colton, Raymond H. (1988). Physiological Mechanisms of Vocal Frequency Control: The
Role of Tension. Journal of Voice 2 208-220.

 

Decker, T. N. (1990). t mentatio : In oducti o S ents ' S ec an
Hearing Sciences. New York: Longnan.
Fairbanks, Grant. (1937). Voice and M'eulap'ep Ml. book (2nd ed.). New York: Harper

and Row, Publishers.

F aure, Marie-Agnes & Muller, Andre. (1992). Stroboscopy. Jamal pf Voiee, 6,
139-148.

Gould, Wilbur J. (1984). The Clinical Voice Laboratory: Clinical Application of Voice
Research. Angels of Otology, fly} ology, and Lamgology, 93, 346-350.

Hirano, M. & Bless, D.M. (1986). Stroboscopic Evaluation of the Lirmx. BostonzCollege
Hill Press.

Hirano, Minoru. (1974). Morphological Structure of the Vocal Cord as a Vibrator and its
Variations. Eolia Phoniatricg 26, 89-94.

Hollien, H., Dew, D., & Philips, P. (1971). Phonational Frequency Ranges of Adults.
Jou_rpal of Speech apd Hemp g Research, 14, 755-760.

Hollien, H. & Shipp, T. (197 2). Speaking Fundamental Frequency and Chronologic Age
irn Males. Journal of Speech and Heam g Research 15, 155-159.

Horii, Yoshiyuki. (1975). Some Statistical Characteristics of Voice Fundamental
Frequency. Journal of Speech and flempg’ Resegrgch, _1_8_, 192-201.

50

Horii, Yoshiyuki. (1983). Automatic Analysis of Voice Fundamental Frequency and

Intensity Using a Visi-Pitch. Journal of Speech and Hgahh' g Reseamh, 26,
467-47 1.

Isshiki, Nobuhiko. (1989). Bhonosurgeg: Iheog; and Practice. Tokyo: Springer-Verlag.

Izdebski, K., Ross, Joel C. & Klein, James C. (1990). Transoral Rigid

Laryngovideostroboscopy (Phonoscopy). Seminars m Speech and Langflge:
Videolamgoendoscopv and Voice Disorders, 11, 16-26.

Kamell, Michael P. (1989). Synchronized Videostroboscopy and Electroglottography.
Jomal of Voice, 3, 68-75.

Kay Elemetrics. (1987). Visi-Pitch/IBM PC Interface: Models 6097 and 6098, Manual.

Linton, Marigold & Gallo, Philip S. Jr. (197 5). Ihe Practical Stah'stician: Simphfi' ed
Bindbogg of Sthtistics. Monterey, CA: Brooks/Cole Publishing Co.

Mackay, Ian. (1987). Phphgtigsz The Science of Speech Exoductioh, Boston: A College-
Hill Publication-Little, Brown and Co.

Martin, Frederick N. ( 1986). Ihtrodugh'oh to Audiology. Englewood Cliffs, NJ: Prentice
Hall.

Nagashima Medical Instruments Catolog.

Peterson, Gordon B, & Barney, Harold L. (1952). Control Methods Used in a Study of
Vowels. The Journal of the Acoustical Society of America, 24, 175-184.

Sataloff, Robert T. (1987). Clinical Evaluation of the Professional Singer. Ear, Nose, and
Throat Journal. 66. 267-277.

Sataloff, Robert T. (1987). The Professional Voice: Part 11. Physical Examination.
Joumal of Vpige, 1, 191-201.

Sataloff, Robert T. (1987). Voice: A New Subspecialty. Ear, Nose, and Imoat Journal,
_6_6_. -

Satalofl", Robert T. & Speigel, Joseph R. (1988). Objective Evaluation of the Voice.
Medical Problems of Performing Artists, 105-108.

Saxman. J .H. & Burk, KW. (1967). Speaking Fundamental Frequency Characteristics of
Middle-Aged Females. Folia Phoniatrica, 19, 167-172.

Stemple, Joseph C. (1984‘). Clinical Voice Pathology: lheory and Mangemght. Columbus,
OH::Charles E. Merrill Publishing Co.

51

Stoicheff, Margaret L. Speaking Fundamental Frequency Characteristics of Nonsmoking
Female Adults. Jo a] o eech and earin esearc 437-441.

Watterson, Thomas L. & McFarlane, Stephen C. (1991). Transoral and Transnasal
Laryngeal Endoscopy. Seminars in Speech and Langpage, 77-87.

Wendler, Juergen. (1992). Stroboscopy. Journal of Voice 6 149-154.

 

Woo, P., Colton, R., Casper, J. & Brewer, D. (1991). Diagnostic Value of StroboscOpic
Examination in Home Patients. loumal of Voice, 5, 231-238.

Zemlin, Willard R. (1988). S eech and Hearin Science: atom d siolo
(3rd ed. ). Englewood Clifis, NJ: Prentice Hall.

BIBLIOGRAPHY

52

BIBLIOGRAPHY

Ad Hoc Committee on Advances in Clinical Practice. (1992). Vocal Tract Visualization

and Imaging. American Speech and Hearing Research Supplemenh 34,
37-40.

Biever, Dawn M. & Bless, Diane M. (1989). Vibratory Characteristics of the Vocal
Folds in Young Adult and Geriatric Women. Journal of Voice, 3,
120-131.

Brewer, David W. & McCall, Gerald. (1974). Visible Laryngeal Changes During Voice
Therapy. Mals pf Otolgyngolog, 83, 423-427.

Cantarella, Giovanna. (1988). Value of Flexible Videolaryngoscopy in the Study of
Laryngeal Morpholog and Functions. Jouml of Voice, 1, 353-358.

Casper, Janina K., Brewer, David W., & Colton, Raymond H. (1988). Pitfalls and
Problems in Flexible Fiberoptic Videolaryrrgoscopy. qumal 91 Voice, 1,
347-352.

Colton, Raymond H., Brewer, David W., & Rothenberg, Martin. (1983). Evaluating Vocal
Fold Function. MW 291-294-

Deal, Randolph E. & Emanuel, Floyd, W. ( 1978). Some Waveform and Spectral Features
of Vowel Roughness. Journal of Speech and Hearing Research, 21, 250-264.

Dempsey, M.E., Draegert, G.L., Siskind, R.P., & Steer, M.D. (1950).The Purdue Pitch
Meter- A Direct Fundamental Frequency Analyzer. Journal of Speech and

Hearing Research 1_5, 135-141.

Gabrielsson, A., Schenkman, Bo N., & Hagerman, B. (1988). The Effects of Diflerent
Frequency Responses on Sound Quality Judgements and Speech Intelligibility.

Journal of Speech and Hearing Research 31, 166-177.

Gelfer, Marylou P. (1988). Perceptual Attributes of Voice: Development and Use of
Rating Scales. Journal of Voice 2, 320-326.

 

Gossett, Claude W. (1989). Electromyographic Investigation of the Relationship of the
Effects of Selected Parameters on Concurrent Study of Voice and Oboe.

Joprpal of Voice, 3, , 52-64.

Gould, Wilbur J., Kojima. H., & Lambiase, A. (1979). A Technique for Stroboscopic
Examination of the Vocal Folds Using Fiberoptics. fichives of Otolaryhgolggy,
L95, 285.

53

Hicks, D.M. & Teas, E. (1987). An Electroglottographic Study of Vocal Vibrato. Journal
of Voice, 1, 142-147.

 

Hirano, M., Kurita, S. & Nakashima, T. (1981). The Structure of the Vocal Folds.

In Stevens and Hirano (Eds), Vopal Fpld Physiology (pp.33-4l). Tokyo:
University of Tokyo Press.

Hirano, M., Yoshida, T., Yoshida, Y., & Tateishi, O. (1985). Strobosfiberscopic

Video Recording of Vocal Fold Vibration. .mals of Otology, M' ology,
_a_r_1d ngology, 94, 588-590.

McFarlane, Stephen C. & Iavorato, Alfred S. (1984). The Use of Video Endoscopy
in the Evaluation and Treatment of Dysphonia. Communicative Disorders 9,
1 1 7-126.

 

Morrison, MD. (1984). A Clinical Voice Laboratory: Videotape and Stroboscopic
Instrumentation. Otolarvngology-Head and Neck Surggrgy, 92, 487-488.

Murry, T. & Doherty, E.T. ( 1980). Selected Acoustic Characteristics of Pathologic and
Normal Speakers. Journal of Speech and Hem g Research 23, 361-369.

Murry, Thomas. (1988). Vocal Tract Parameters Associated with Voice Quality and
Preference. Journal of Voice, 2 , 111-117.

.\'lyers, Denise & Michel, John. (1987). Vibrato and Pitch Transitions. Journal of Voice
1, 1 57-161 .

 

Peppard, Robert C., Bless, Diane M., & Milenkovic, Paul. (1988). Comparison of Young
Adult Singers and Nonsingers with Vocal Nodules. J oumal of Voice, 2,
250-260.

Rontal, E., Rontal, M., Jacob, J .H., & Rolnick, M.I. (1983). Quantitative and Objective

Evaluation of Vocal Cord Function. Appals of Otology, R_hm' ology, apd
Lagmgology, 92, 421-423.

Saito, S., F ukuda, H., Kitahara, S., & Kokawa, N. (1978). Stroboscopic Observation of
Vocal Fold Vibration with Fiberoptics. Folia Phoniatrica 30, 241-244.

 

Saito, S., Fukuda. H., Isogai. Y., & Ono, H. (1981). X-Ray Stroboscopy. In Stevens
and Hirano (Eds), Vocal Fold Physiology (pp. 95-103). Tokyo: University of
Tokyo Press.

Schmidt, P., Klingholz, F.. & Martin, F. (1988). Influence of Pitch. Voice Sound
Pressure, and Vowel Quality on the Maximum Phonation Time. Journal of
Voice, 2, 245-249.

54

Walker, J .S. (1988). An Investigation of the Whistle Register in the Female Voice.
10mg] of Voice, .2, 140-150.

Watterson, Thomas L., McFarlane, Stephen C., & Brophy, John W. (1990). Some Issues
and Ethics in Oral and Nasal Video-Endoscopy. Seminars in Speech and

Langgge, 1-7.

White, James F. & Knight, Robert E. (1984). Oflice V ideofiberoptic Laryngoscopy.
Lgmgoscox, 94, 1166-1169.

Williams, G. T., Farquharson, I. M., & Anthony, J. (1975). Fibreoptic Laryngoscopy in
the Assessment of Laryngeal Disorders. e Joum of 010 nd Otolo
_82, 299-315.

Yanagisawa, E., Casuccio, J.R., & Suzuki, M. (1981). Video Laryngoscopy Using 3 Rigid
Telescope and Video Home System Color Camera - A Useful Office Procedure.

Annals of Otology, Rhinology and Laghgoscopy, 90, 346-350.

Yanagisawa. Eiji. (1982). Office Telescopic Photography of the Larynx. Ahhals of
Otology, Rhinology, and Lag'ngology, 91, 354-358.

"‘lilillliillfliill“