close

Вход

Забыли?

вход по аккаунту

?

Studies in the Pitch and Duration Characteristics of Superior Speakers

код для вставкиСкачать
STUDIES IN THE PITCH AND DORATI ON .CHAHA0TEBISTICS
OP SUPERIOR SPEAKERS,
by
John G« Snldecor
A dissertation submit ted la partial fulfillment ot the
requirements Tor the degree of Doctor of idxilosophy,
in the Department of Speech, in
the Graduate College of the
State University of Iowa*
August* 1940
ProQuest N um ber: 10984085
All rights reserved
INFORMATION TO ALL USERS
The q u a lity of this re p ro d u c tio n is d e p e n d e n t u p o n the q u a lity of the co p y su b m itte d .
In the unlikely e v e n t that the a u th o r did not send a c o m p le te m a n u scrip t
and there are missing p a g e s, these will be n o te d . Also, if m a te ria l had to be re m o v e d ,
a n o te will in d ic a te the d e le tio n .
uest
P roQ uest 10984085
Published by ProQuest LLC(2018). C o p y rig h t of the Dissertation is held by the A uthor.
All rights reserved.
This work is p ro te cte d a g a in s t u n a u th o rize d co p yin g under Title 17, United States C o d e
M icroform Edition © ProQuest LLC.
ProQuest LLC.
789 East Eisenhower Parkway
P.O. Box 1346
Ann Arbor, Ml 4 8 1 0 6 - 1346
T \ W
bt/1
15a# writer wishes to express M s appre­
ciation to Professor &r&nt Fairbanks for his
suggestion of the research problem® and for
valuable guidance throughout the progress of -the
investigations*
Hi© writer also wishes to thank
Professor Edward 0# Mabi© for M s interest and
aid during the period of graduate study#
Special
thanks are due to Brs* W* L* Pronovost, G. w#
Mg Intoall, Jr*# and IS* T* Gurry for their assist­
ance and cooperation In gathering part of the
data for these studies*
lit
TABLE OF CONTENTS
&
c o m p a r a t i v e ; STUDY
OF t h e p i t c h a n d d u r a t i o n c h a r a c t e r i s t i c s
OF IMPROMPTU SPEAKING AND ORAL READING
Pago
XOTROOT GTX OH
1
KXPERXMEOTAk PHOOTDORK
1
rbsum s
6
1*
.K W h £>®wX
6
2*
P tto la i V a ri& M X X fc x
9
3*
Mean B te to n t o f B io n & fcX e n a j
15
4#
3.
E a t® o f Pifc® & Chang® a n d Sfuaaber
" -o f T ffc W f fia S io s M
^
20
R a to
23
mmma
25
PART II
AN OBJECTIVE STUDY OF PHRASING IB IMPROMPTU SPEAKING AND
ORAL READING
Page
INTRODUCTION
27
EXPERIMENTAL PROCEDURE
28
RESULTS
32
1,
The Location and Number of
2*
The
*KO
i
Ji-j
_of the Percentage
b"~£he"Durat ton o.
34
XT
Fag©
5*
T m Donation
tmmi#©
0 Of
4)
59
III*
m
SUMMARY
47
jmmTXQA^lOM OF TUB PITCH AHD DURATION OTARAOTBRXSTIOa
OF THE ORAL HEAP1H0 F3RP68MAB0B3 OF SUPERIOR WOMEH
SPEAKERS
Pago
I*
II.
IHTROOTOTIOH
50
EXFEEIHEHTAL FROOBDUSB
51
RESULTS
54
X.
54
Pitoh Leval
of Fifseh
61
Hionations
65
4*
Rato of Finite
"of
awa
6*
71
of Bate
73
Pupation of Phon&tXons, Faaaooj
76
7,
Itofo.ey of InfXootXons and Shift©
8,
.iwf
t*p©w^ WFoapooutage
€&aoa
within
I
unwi
»
■of
■&&>» i
ti'riMt^^w iwiwiii- P
jji'Mn'p.mwrwiiwuaii
76
X h S y w l ■Frontne
76
9t3UCftSS:
79
V
APPEHOIX TO PARTS I AHD IX
APPS3SDXX TO PAHT XXX
82
TssbXo I .
Table XI.
29
. . . . . . . . . . .
Table 111..............
Table XV.
Table V.
Table VX.
Table VII,
33
36
............. .37
.........
. . . . . . . . . . .
38
41
....................42
Table VIII.........
44
F&gurfc I*
11m
# * , . * * * * , * *
56
* * . ......... . * *
m
XfSlljiX© XU*. ^# # * • * * # # # #
60
MfeX# XV*
• *##*.*****
62
***********
64
^StlbX® V#
*####*#*,***.
66
» i t VI*
***********
68
$«&1* ¥11* * * * * * * * * * * *
¥0
ttt&ft ¥1X1* * * * * * * * *
¥2
J%g£tr* XI#
Tafcl© XX*
X*
fsteX© XX#
# ^.
***********
¥4
***********
¥5
..* » ♦ » * . , . , * * *
77
The University of Iowa
LIBRARIES
PAR* I
A. Q O m m A T t V E SfTOT OF THE FITCH AMD DUEAT IOH
CHARACTERISTICS OF IMPROMPTU SKBAEXH-0 AMD ORAL BSAPIMG
Recent inv© s11gat Ion s by Fronovoat^, and McIntosh^
X*
2*
Pronovost, W#, An experimental study of the Habitual
and natural pitch levels of superior speakers* Ph*D*
Dissertation, State University of Iowa, 1959*
MeXntoah, G* W* Jr., A study of th© relationship be**
tween pitch level and pit eh variability in the voices
of superior speakers# Ph.D. Dissertation, State
DniVersity of Iowa* 1939#
have described objectively the pitch and duration character­
istics of th© oral reading performances of six carefully
selected superior male speakers.
The purpose of the present
investigation is to study, under like experimental conditions,
impromptu speeches and subsequent oral readings of these
.speeches by the same subjects.
X*
BXFEHIl^KMTAh PROqgPTOK
Th© subjects used in this investigation were th©
same as those previously selected for the research studies
mentioned above*
They were chosen from a group of 25
superior male speakers, recommended by the staff of th©
Department of Bpeedh at the State University of Iowa for
2
their super!©r voice ns&go#
A rigid, teat of speaking and
tiJra& reading ability ©iiMn&fced from consideration all
who presented deviations in articulation and voice usage
in either speaking or reading*
this preliminary test*
Eight subjects survived
Phonograph recording® were made
of four readings of a carefully composed prose passage
for each subject*
She recordings were mad© in a sound­
proofed room* and the subjects were instructed to- read as
if to an audience of 25 people*
The best of the four
readings for each subject was selected by seven trained
observers who ranked the performances in their order of
general effectiveness*
Twelve trained observers then
ranked these eight best readings by the method, of paired
comparisons in which each reading was compared to each
other reading twice, once preceding and once following*
fhe pairs were presented in random order*
On the basis of
these judgments, two more speakers were eliminated, leav­
ing six subjects for the studies.
It will be noted that
the superiority of these speakers as impromptu speakers
was not specifically demonstrated*
However, the general
superiority of their speaking ability may be safely assumed
on the basis of the criteria by which they were originally
chosen* and the additional selective processes based upon
their oral reading performances*
For the purposes of the present study, a strictly
impromptu speaking situation was arranged in the following
manner*
Each of the six subject© was given the topic
*Hjr future Job,n and allowed om
minute to prepare a talk
which in tern was to be delivered in approximately one
minute.
The subject was taken into a sound-proof room
and naked to apeak ms if to' an audience of 25 people*
fbe entire speaking-performance was recorded with highquality recording equipment*
One week later the subjects were called back
into the same experimental situation In order to record
a reading performance directly comparable to the impromptu
speaking performance*
Each subject was given a typewritten
copy of M s original speech, which had been previously
transcribed and punctuated In a simple fashion, and allowed
one minute In which to prepare himself to read it*
He was
then instructed to read as If to an audience of 25 people,
and the performance was recorded*
Chiefly because of the time involved in dealing
with frequency measurements, it was necessary to sample
the performances*
As a precautionary moans, the intro­
ductory and concluding sentences were eliminated*
relatively complete thought units were used*
Only
The speech
of Subject 0, with the sampled portions in italics, is
presented below, to indicate the general nature of the
performances and the sampling procedure *
One of the most important things with
which all of us have to contend, of course,
Is preparation for ‘the future job which each
one of us will hold when we get out of the
university* In my own case the particular
preparation which 1 am to make, and which I
think everyone of us could make in prepara-
felon for a future job* As, first of all,
very careful preparation in the ability
to meet people* And that of course come®
praofeMt; in irfe^fIue people* ffie
is BSSst 33P-r
m,»iwi<wfiwiib-ji»o
®FpEE1
US;332U1
.
©gr c w r & e mv^.lyee feajtsin
eel
i
m
e
y
,r1’
'anci'
very c.a#cJ^Tx^
,
il
oulSieaa ori^nACatioh Ia
,,r xEe felilrd principle m i n g wElcH Is
I^orfeant in preparing for a future job,,
in iay ease as well as in the ease of each
one of yon,. Is taking ear© to create a
liking between yourself and the people
with whom you associate*
In duration, sample® of the speaking, performance
range from 23• 31 to 26*88 seconds, and those of the read­
ing performance from 17*85
from
to 24*12 seconds*
ffliereare
53 to 84 words In the selections studied, with slight
discrepancies in the number of words of the two perform­
ances because three of the subjects omitted from one to
three words in the reading*
A total of approximately
24,000 individual sound waves were considered for the entire
study*
The phonophotograph!e technique originated by
*# modified by Simcnft* and Lewis and Tiffin5*
3*
4*
5*
Metfeasel, M*, Techniques for the objective study of
Vocal art, Psychol* Monog** 36, 1927, 1-4Q*
Simon, C« r
f ,1
“ftierv a'rlab'ility of consecutive wav©
lengths in vocal and instrumental sounds, P
36, 1927, 41-83*
Lewis, D. and tiffin, J#, A psychophysical study of
Individual differences in speaking ability, Ar* Bp*,
1, 1934, 15-60*
“
Pitch
Level (©*$>* s+}
Median Pitch
Levels
IS ©
17*30
Mean Pitch
Level (s*p.s«)
Mean Pitch
17*30
18.19
Mean S* D.
(tones)
1.32
1*65
Mean Total Pitch
Range (tones)
S.42
9.89
eTosues above 16.55 cycles per
aeeond*
and electrified by Oowanfo was used*
©.*
The use of this
®ovimt M#, Pitch and, intensity charae t©rl stic a of
stage speech, Ak> 8g*.» X, Buppl*, 1936*
technique gives a photographic record of fundamental sound
wave frequencies from phonograph records.
In the present
study, the average period length for consecutive intervals
Of 0*038 seconds each was measured*
The over all error
with this method of measurement is negligible, being of
the order of 0*5 per cent, or 0*04 of a tone*
Frequency conversions were made from period -1 ength
measurements, graphic pitch scores plotted, and various
computations and mean® of analysis applied to the data,
following procedures outlined by Fairbanks*7
7* Fairbanks, G*, Recent experimental investigations of
vocal pitch In speech, *T*A*J>.*A* 11, 1940, 457-466*
11*
1*
RBBXT&fS
Bitffh Level* Table I presents data on group
measures of central tendency and variability for the two
performances studied*
Measures of central tendency are
stated In the first four items of Table X*
It will be
seen that the median pitch level for Impromptu speaking
is 17*30 tones, and that for reading, 18,00 tone® above the
*«-o reference frequency®, a difference of 0.70 tones in
7
C4
261.6
C3 -
129
146
145
130.8
120
’2
65.4C o m p o s it e
B
D
- READING
- SPEECH
c,
32.7
distribution of pit olios
$ $ & £ & tm d ~ e a d ln ^
re p ro a o n '
io n ta ;
8
»«ww«#»M— m ww mHmmr-m****' ..iw ■,m«ii
8*
S&e a»jPO ffefersno® f3?®<suienoy proposed by Fletcher, H.,
Louclnoaa, pitch, «Ad tlasbre of nSAfsioal tones, £,A.3.A.
6
,1934,50-69.
favor of the reading perforates#
In the case of the- mean
pitch levels, the reading® are 0*89 tehee higher than the
speeches, a'difference very similar to that between the
median pitch levels#
When the vertical relationship between
those items In the table Is considered, it will be noted
that the median and mean pitch levels are alike in the
ease of speaking and show a negligible difference of 3 e«.p#s*
in the reading*
In the individual distributions the mean
pitch level® fall slightly below the median pitch level in
eight out of the twelve oases#
This tendency for a alight
skewing towards the lower frequencies is character!atie
of distributions of- pitches for the speaking voice#®
w*1
9*
ww*1
iiKW*iH*iww*!»#i»«M**wwNw*EW*wpweMjiw<rta*tCwqjwt»^
See Hclntosh^ f^opove®tA. op#. pit#
Figure 1 pictures the distributions upon which
these computations were based#
Musical tones are plotted
along the ordinate, and each abscissa is the percentage
of total pitches used#
The distributions on the left
side of the figure represent the composite of all pitches
for the six subjects*
The remaining distributions, from
left to right, are those for the individual subjects in
descending rank order of median pitch
x
9
levels &os? the t wo perfo t m m m ® * From this figure, one may
see tlie differences between the median pitch levels for the
#*mpweltoe distributions already discussed*
In addition.
It will be noted that individual differences between the
medians for five subjects tend to be alight*
For Subject A
the median pitch levels for each performance are Identical*
Although reading has a median pitch level only
0*70 tones higher than that for Impromptu ©peaking, this
Increase of a small order represent a approximately the
m a m difference as that- found by MoXntoahJLO between the
10*
MeInto shg» Q* W* %Tr.#, op* oft*
median pitch levels of normal readings and readings at a
slightly higher pitch level*
HI® study indicated that
flexibility increased as pitch level was raised#
Further
measures In the present study will Indicate that impromptu
speaking is less flexible in pitch than the readings of
the speech*
However, it should, not be presumed that this
slight increase in pitch level 1© the sole factor con­
tributing to the additional pitch flexibility of the read­
ing performance.
2*
Fitch Variability*
Has mean standard
deviation in tones is presented as the fifth Item In liable
1*
Hie mean standard deviation of the individual distribu­
tions is used- in this case rather than the standard
10
Table XX#
Increase of
m m *m m mti&iw m'm » i m m w
m m m *
W m m x m over
RmMtm
10
*52
*59
*07
20
*62.
*78
*16
50
1*05
1*21
*18
40
1*45
1.65
*18
50
1*79
2*15
.56
60
2*17
2*71
*54
70
2.66
6*5-5
.69
SO
6*25
4*04
*81
90
4*02
8*19
1*17
iocte
8*42
9.89
1*47
45-DIstributes equally above and below the median.
^Equivalent to total pitch range#
deviation of the composite distribution* in order to avoid
Che leveling effect from combining distributions that have
different central tendencies*
Impromptu speaking with an
$*D* 'Of 1*52 tones la seen to bo less variable than oral
reading of the speech with an $*$* of 1*65 tones*
Standard
deviation® for the individual distributions were in all
oases greater for reading than for speaking*
The standard
deviation as a measure of pitch variability#, or flexibility#
has been validated by McIntosh^, who-found that in each
U*
McIntosh*. C. w* Jr** eg* ejt*
ease this measure was larger for normal readings and read**
tags more flexible by instruction* than for readings which
were less flexible by instruction*
In fable II will be seen the mean values for thy
extent in tones of median percentages of the pitches used*
fhe computation of these values m y be explained as follows*
ISaeh individual distribution of pitches was considered*
and the median 10 per cent* 20 per cent, etc** of the oases
computed up to and Including the median 100 per cent* which
last measure is equivalent to the total pitch range* or
the distance In tone® between the lowest and the highest
pitches used*
$he median 10 per cent m y be considered as
the rang# between the 45th and the 55th percentiles* the
median 20 per cent as the range between the 40th and 50th
percentile®.# -etc#
Tbm value of this computation will be
IS
JtotoA when lb is observed that the medl&m 90 per cent of
the ease© in reading is subtended by only 5*-19 tones,
whereas the total, rang# is i*.&0 tones,. m. increase of
4* TO tones, when cmly 10 per sent more of the extremely
high and low pitch©© are added to the distribution,
in*
spscfelon of the tttall©” on the distributions in Figure I
will also Show how a few Isolated pi tehee increase the
pitch range out of proportion to its true wains as a
measure of variability*
From fable IX It will be noted that reading is
mere variable than impromptu speaking at all percentages.
The greatest difference between the two performances will
be seen in the median 90 -and 100 per cent measure®, in
which there is a difference of 1*17 and 1,47 tones,
respectively*
Further Inspection of the table, from the
40th to the 90th median percentage measures, indicates
that a given rang© in speaking subtends approximately 10
per cent more of the pitches used in such performances
than the same range include© in the reading performances#
the mean total pitch ranges for Impromptu speak­
ing of 8,42 tones and for reading of 9*89 tones are
greater than those reported by Iteray ..and Tiffin^,, who
12#
Murray* E* and Tiffin, J*, An analysis of some basic
aspects of effective speech, Ar* £g*# 1, 1934, 41*83*
found pitch ranges of 2*5, 4*5 and 8*3 tones for poor,
good, and trained speakers, respectively#
15*
Lynch3*5 reports
Lynch, 0* E*, A pho t©phonographic study of trained and
untrained voice© reading factual and ©motional material,
iS* SSL** 1» **34#
pi tola ranges of 5*8 and 8*8 tones for untrained and trained
speaker©*
14*
Lewis,,-and ,fiffin^ report pitch ranges from
Lewis, D* and tiffin, J*, op# eit#
3*8 to 0*0 tones*
Their highest value of 0*0 tones Is
about the same a© that for the average total rang© in
reading found in this study, hut much lower than the
pitch range of 14*18 tones found for the reading performance
of Subject E in this Investigation#
It is interesting to
note that the range of this on# subject would become even
greater If augmented with frequencies in his performance
which extend as low as 81 o«p»s*
However, no frequencies
lower than approximately 47 c*p*a* were included,, because
they occurred only momentarily and varied markedly from
wav© to wave*
The presence of such low frequencies in the
human voice 1© attested by Dudley3*5* who found It necessary
15*
Dudley, II*, Remaking speech, monograph B-1171,
Bell telephone System Technical Publication, 1939#
to use a frequency discriminator in the "vocoder" to
ON
WANT
TO
GO
---------------------- ^
1
.f
t
UJ
UJ <
1- W
1
UJ I i
S l-<
.
i
>
.v /
/ *’’’
<*
11
11
u
Jf/
l
BE­
LIEVE
THAT
THE
V i\ V \ i'1 1I
1
I v/
, »
rtti
$i A
i
1
\
[\
SO
\
!/*
1/
AND
,
v
)
______ ir -^ . _____
_
\
1
BE­
LIEVE
SO
AND
EVEN­
TU ­
ALLY
STAGE
THE
ON
WANT
TO
60
^ 1
-*
iu .
E 1
11-
z
/
l
15
•Uatetti ntroublesome pxftn* below 50 e*p#a*
For both of the measures of pitch variability
4tfM U M « d above, impsraeptni speaking is loos variable
torn W & X reading of the ©am© material.
Without exception,
the value© for both speaking tod reading.are less than
those reported by McIntosh^ for;a normal reading of a
15*
McIntosh, 0, W, Jr*, ©gu. sit
well composed prose passage by these same subjects.
$he nature of the performances and the type of
material used m a t be considered when interpreting these
differences in pitch variability*
In the first place,
impromptu speaking is not the type of performance in which
to expect the same sureness of approach usually anticipated
in the oral delivery of1well written material *
Th±®
probable lack of sureness may be reflected in the relatively
mediocre composition of the speeches*
Secondly, because
these speeches were read almost word for word as they had
been spoken, it seems unreasonable to assume that even the
best oral readers have the ability to surmount the limita­
tions imposed by the weaknesses of the original performances*
3* Mean latent of. Ihonationa, inflections, and
ghlfts* Figure 11 presents superimposed pit to curves of
Subject a for the first few words of M s impromptu speech
and reading, with the speaking shown by a solid^lin© curve,
and the reading by a dott^d-lin© curve*
T im ordinate
16
Indicates the musical scale, and one second Intervals of
time are marked along the abscissa#
From such a curve
one may measure, (1) the extent and duration of phomtions,
(2) the extent and duration of inflections and their
relative upward or downward direction* and (3) the extent
of shifts*
A definition of these objective measurement©
is stated by Fairbanks!^
XT#
The .measurement of shifts between
Fairbanks* &*, op* eit» tr*#•* **.*•*&n inflection is
defined as a frequency modulation,: wh e W e T upward or
downward* without'interruption of phonation, while
th# term shift refers to a change in pitch which take©
place between the terminal pitch of a given phonation.
end the initial pitch of the subsequent phonation*
13he rate of pitch change i# a measure of the rapidity
with'lffiSen frequency "!© "modulated per unit of time
during inflections, 1* e*, the relative "steepness11
of the inflections* For any inflection this is
determined by dividing its extent la tone© by its
duration in seconds* A change in. direction of pitch
movement is a shift in"frequency mediation from mx '
ujSRSraT^reeticm to a downward direction, or vice versa*n
phrases was postponed until a usable technique for indicating
the phrase limits could be developed*3*®
18#
See below, fart XX*
Figure 11 serves to illustratet
(1) marked
similarities in the general pattern of the two performances,
(2)...measurable difference® in flexibility, and (3) an in­
crease in duration during ix^romptu speaking*
Differences
in median pitch level are not indicated in this sample,
±7
f a m © XXX*
Piff,
Mean Extent of Hhonaiiona
2*54
2*80
*16
Upward Inflections
1*67
1*92
*28
Downward Inflection©
1*60
1*82
*22
All Inflections
1*65
1*86
.23
Upward Shift©
2*03
2*18
*12
Downward Shifts
1*84
1*68
*» *16
All Shifts
1*96
1*98
*02
Mean Extent of Inflections
Mean Extent of Shifts
18
tout are demonstrable when larger ©ample© are considered*
✓ !®tol© III H a t e the mean, extent© of phonations,
Inflection© and shifts*
Ihe mean extent of phonatlona
■In epeajcing la seen to to© @,*34 tone©, or 0*18 tone© leas
than the value of 2 *50 tone© in reading*
Con siatent with
previous trends, the mean extent© of Inf lections are less
In the ease of the speaking performance for upward* clownward end all inflections*
$kmn the mean extents of clown**
ward end upward Inflection® are compared, the latter are
slightly greater In extent in tooth performances*
This is
the reverse of trends previously reported elsewhere, and
some speculation may toe in order*
It has been quite common­
ly stated that rising Inflections Indicate an Hncompl©te­
nses1* or ®lack of sureness” In vocal performance*
In this
Case it may to© that impromptu speaking tends to to© an
nunsur©w performance, and reading of the speech invites a
us© of similar upward inflections that are greater in extant
than downward Inflections*
in keeping with the above
findings, the mean extent of upward shifts and of all shifts
In reading are seen to exceed those in speaking*
The mean
extent of downward shifts, however, 1® greater In speaking
than in reading*
Mien upward shifts are compared to down­
ward shifts for tooth perfomances, it 1© seen that upward
shifts are greater in mean extent#
Mien the extent of all
inflections and all shifts are compared, the shifts are
seen to exceed the Inflections in mean extent toy one-third
of a tone*
Hit© last finding is in opposition to previous
19
fable I?*
fessures of r.fote Ojf x&tdh change and mrnber of
tup < S f g S s a t e f e
J8M r i 6^ C
i g g g l « ^ g
Biff.
a #02
10.86
2*84
411 Infleetions
13*40
16.88
5*48
Downward Inflections
18*38
15.74
8*1©
Upward Inflections
14.61
18.81
5.90
ail Extent®
*7.00
7*05
.05
Semi-tone or Greater
4*20
4.65
.45
Dess than a 3©mi-tdtt©
2*80
2.41
S!
*
1
6.25
7.05
.78
Semi-toil© or Greater
O
o•
*0
8.18
1.18
Dess than a Semi*ton©
2.25
1.85
—.40
Mean Bata of Pitab Changes
Mean Bata of Pitch Change.
Baring inflections
Mean Bomber of Changes in
Direction of Pitch per
second:
During Piionatlon Only
Including Shifts
All Extents
« f W B per second*
/
reports whlOh have indicated that inflations exceed
shifts in moan extent*
With the oxooption of downward shifts# those
measures of phosmtions> Inflections and Shifts are greater
in rea&Atg than in speaking*
Shifts exceed inflections
in mean extent*
of Pitoh Change and Ifapber off Pitch
The fj.rst four items in Table IV represent
«
Fairbanks, G#,
the composite rate of pitch change for the two performances*
It will be observed In fable IV that reading
exceeds speaking In mean rate of pitch change* in tones
per second, for the entire passage* and during all In—
flections, downward Inflections and upward Inflections*
I d s difference Is relatively large* the actual percentage
of Increase exoeediij^ 25 in all cases*
It appears that
impromptu speaking and oral reading of the spa aches are
clearly differentiated on the basis of differences in
the mean rat© of pitch change between the two performances*
Table IV also lists data for the mean number of
>•
changes In direction of pitch per second* Inspection of
lower part of this table Indicates that changes In
direction are more frequent in reading than in speaking,
with the exception of those changes of less than a semi-tone.
tu toes# tottor measures the Changes are more frequent in
epeaktog than in reading#
The differences in favor of speak-
tog daring phonafcton only are very small#
When shifts are
included, to© inorea#© 1® relatively greater, with the
exception noted, hut net extreme*
In general, it would be
expected that reading would hare more changes in direction
of pitch per second because of the feet that the reading
performance la much more rapid,: and yet contains approximately
the same contextual mat ©rial.
On the other hand, it is not
surprising to find that the m e m number of changes of less
than a semi-tone, Including shifts, are very slightly more
frequent to speaking, for, although there ar© about an
equal number of shift© In both performances, there are
relatively more shifts of a email extent In speaking.#
^InfcpshffQ found that there were more changes In the
<m infr»’i» w » w » n « w w .«i ■■nw n >v »wp«w " < » " i'i 'i ?m; ■’»i»—ii«ih x >
20#
ih iiiwhumiwb m i. m i ■
McIntosh, 0* W* Jr.#, op# clt#
direction of pitch far the mare flexible performances,
and to general this trend Is supported here#
v-..
The relative numbers and. percentages of Inflections'
and shifts to this study are not considered In detail
because of the fact that the performances differed consider*
ably to length and context and therefor© varied in the
numbers of these
and toflections#
However, when the
relative member* of these *mepsurefe; are 'Conkidereci and t o m
Speak-
Composite Hate
(word® per *aln#)
tost
151
Head«im*w«i—mi»
toR
Dlff#
niiii»nwiwr
183
32
Imrease
■
>»f^eHrtwi
wf
r
iw,i■
or Deerease
81
Mean. Ratio of
Phonatad Time to
Total Tim®
#66
#68
#03
5
Mean Duration of
PhonetIons (see•)
#38
.35
-.03
-8
Mean Duration of
Pauses (see*)
.80
#16
-*04
-20
averaged# there «*« approximately S@*
andfiv© per oent
Of each inflection® far downward# upward and level In-*
fixations* respectively.
When shift® within phrase® are
Considered, there are found to bo approxii&at©ly 25# 55#
and B2 per cent of downward, upward# and level shifts#
respaet±vely # in both speaking and reading.*
toe l^oromptu
speeches did not differ substant± ally from the oral reading
perfoManee®.
In general# there are definitely more down­
ward inflection® and upward shift® in both performonee s#
and the value® indicated here are approximately the same
a® too®® reported by l^.lntosh^* except that in the ease
21*
BeInto#i# Cf W, Jr«# op.eit*
of the present study there are relatively more level inflections.
5*
Hate*
In Table V# which- presents ooi^ositc
measure® of time for both performances# it will be seen
that the impromptu speaking performances were delivered
at a mean rate of 151 word®, per minute# whereas# the mean
rate of the reading performance® was 183 words per minute.
This difference is seen to be a striking one when these
rates are comp&red to Parley*
22
norm® established for a
* Parley# F* P., A normative study of oral reading rate#
M*A* toesia# State University of Iowa# in progress*
24
random sample of' college freshmen reading factual material
aloud*
lb© spcaking pcrformiuacc lies between tha 10th and
15th percentiles and the reading of the speech between the
80th and 85th percentiles for hi# subject## Obviously,
however, any comparisons made with these norms would appea#
'to bo valid only for the reading of the speech.
Because
there are no known no^Pm# if©«r rate <Ln linpromptu sp>©£djtir'g,- It
is Impossible to say at what percentile for this typo of
performance this value falls*
In considering the Individuals*
all of the subjects showed a more rapid rate in reading
with the exception of one subject who read and spoke with
exactly equal rate©#
Hxe length of word# in these perform^
ances Is approximately the came m
that for the passage
studied by Parley#
25#
Parley* IM*, og* pit#
fhe second row of !EabX© V show© that total
phonated time represent# a relatively constant fraction
of the total performance time for both performances* in
spite of the marked difference# in rat©#
This is possible
because pause© and phonatlons are seen to consume more
time In the Impromptu speech than in the reading of the
speech.
Because of the general increase in rate in the
reading performance* it might be predicted that the mean
duration of pauses would decrease relatively max’© than the
mean duration of phonatlons.
Inspection of liable V indicate#
25
tefct such mm assumption &© true, a© the percentage of
decrease for th© duration of phon&tion® la seen to bo eight,
whcre&c teat for tea duration of pauses is 30 per © ©nt*
in.
m m m w
Phonograph recordings were mad© of six superior
speaker® delivering on© minute impromptu speeches*
Om
week later the subjects returned to the laboratory and a
second set of recordings were made as each subject read
aloud a copy of hi® speech#
Hionophotographio techniques
were utilised to obtain records of the pitch and time
©haracteristic® of these performance©#
te© application
of various measures provided description® which contri­
buted to the following general conclusions.
1*
tee reading© of the speeches were slightly
higher in pitch level than the improas^tu speeches, with
the exception of the performances of one subject , whose
pitch level was identical in both performances.
2m
Heading was found to exceed impromptu ©peak­
ing in pitch variability in all computations#
5#
With the exception of downward shifts,
phonations, inflections and shift® were greater in mean
extent in reading than in ©peaking.
4#
Heading exceeded ©peaking in mean rate of
pitch change.:, and In the number of changes in the direction
of pitch per second^ with the exception that the mean
26
m m her of changes in pitch was greater in speaking* when
changes of loan than % semi**tone wore considered and pitch
shifts Included#
&t
Heading* with a rate of 185 words per minute,
was notably more rapid than speaking* which had a rate of
151 words per minute*
S*
She proportion, of phonatecl time to total
time was almost identical in both performances.
However*
there were more phonations and pauses of longer mean
deration in speaking than in reading#
?•
In general* subject to the limitations of
the experiment*, reading of the impromptu speech was higher
in pitch level, evidenced more pitch variability, had a
more rapid rate of pitch change, more changes in the/
direction of pitch, more rapid rate, fewer pauses and
phonations, Shorter mean duration of pauses and phonations
than did the Initial performance Cf the impromptu speech
itself'*
m
o b w o t e sm m
of phrasing in
1MFH0MFTO a r m m m aid deal readxno
Measurement of the so-called flphi»asesH of the
impromptu speeches and oral reading performances studied
above was considered desirable in order to complete the
descriptions of the pitch and time Characteristics of
these performances#
However, upon examination of the
objective records and the textual mat ©rials of th© speeches
and readings* it was found impracticable to make any
arbitrary selections of units which might be called "vocal
phrases*w
in th© first place, the performances of the
different subjects varied in text, in number of words, and
in length#
Secondly, the impromptu speech of each subject
differed from his oral reading of the same material with
respect to th© location and duration of pauses, the rate,
and, to some extent, the general movement of pitch.
Because
no statistical procedure appeared to provide a solution to
these difficulties, th© possibility of arriving at th©
objective characteristics of phrase© through th© judgments
of trained observers suggested itself as a satisfactory
solution to th© problem*
2 8
1
Eeeord# of toe previously studied impvowta
apeetoes « M reading# were pimp*# before twenty advanced
gradu&t© etudenb® in toe Bep&rtmenb of Speech at toe state
totveraity of l m «
these observer# ware mqpplied. with
wnganurfcuated eopies of both perfomamee# typed in capital
letters* and wore asked to indicate with vertical line®
those points which* in their ofinio% appear to limit to#
phrase*
Bo restrtetloms on time war# imposed#
site
following direction# war# given both verbally and in typed
fm#
W© are asking yon to partloifat# in a
procedure which wo hop# will amble us to
define a spoken phrase>objectively* Instead
Of asking you for a verbal, definition* we are
asking yOm to define by example*
are famished with records of speak*
ing performances and oopies of toe material
recorded* Flay the record# as many times ,as
you wish# tedieating how you think toe material
ii» phrased by too speaker by Indicating toe
limit# of each phrase with clear vertical lines*
order that this experiment will be
valid# you must disregard how you think toe
ejection,j|jgBfl&»+ t
phraaln^as
e
t
o
f g ^ t^ele^rds*, Bo
In order to illustrate toe nature of toe phrases
indicated- by toe## observer## and toe results of these
Judgment#* a copy of sample performances for one subject
is presented in Figure 1*
Phrase limit#! are indicated
29
fable tf
lean
P e re a B ia g ® o f
Observers
5
2.0
1©
SO
v 2©
30
36
40
46
60
6©
60
66
70
7©
SO
85
@0
05
100
on of
as jLimltlm
Bumber of
ftoMO Idmlts
17
18
17
8
7
8
4
5
6
4
4
10
0
6
4
4
•8
6
15
50
Me&m Doratio b
of Pause® (sec *)
♦05
♦06
♦04
*03
*04
*07
*0©
•12
♦15
*15
♦27
,29
*31
#14
•59
*26
•54
*89
*53
*73
30
1*
Her® and ©isewher© In this .study the tern phrase limits
is used to refer to those points which w e r o ^ S a g e T ^ s "
limiting phrases*
as follows t Hie number a in parentheses above and below
■the lines In the text represent for Impromptu speaking and
reading., respectively* the percentage of observer© who
indicated that these point®, in their Judgments* limited
vocal phrases In the performances of Subject 0.
Similar
tabulations were made for both performances for each subject*
Measurements were then made from the objective record of
those point® which had been judged as phrase limits, and
of the phrase® which fell between these indicated limits*
fable I Indicates th© relationship which exists
between the percentage of observers who marked phrase
limit® in the texts of the two performance®* and the mean
duration of the objective pause® at these points*
When
this table is scrutinised it at once becomes apparent that
an increase in th© percentage of judgments is accompanied
by an increase in the mean duration of pauses*
It Is also
seen that th© mean duration of pauses accelerates rapidly
where th© Judges agree more than 60 per cent of the time*
Although reversal® occur* the progression would probably
smooth out If th© sample were increased*
Because of this
acceleration it was deemed desirable* for purposes of the
objective measurements of the present study* to divide th©
phrase limits at the 50 per cent point* comparing these two
(S)
(6)
THEREFORE
Hr PREPARATION FOR IT
(S)
I WOULD TOY
{30)
(S)
TO GET
(100)
EVERY ANGLE
(60)
OF PRODUCTION
(5)
AND
{15)
(100)
DIRECTIOR
OF PLAYS
(95)
THIS WOULD ENTAIL
(JS)
COURSES IN SCENIC DESIGN
(100)
COURSES IN STAGE
(R)
(100)
LIGHTING
(40)
AND THE OTHER TECHNIQUES
OF
(100)
PRODUCING A PLAY
(95)
IT WOULD ENTAIL NECESSARILY
(E>
(S)
(85)
ADVANCED COURSES
(R)
(3)
APPROACHING
(R)
(S)
A PLAY
(R)
IN DIRECTING
AND IN
(0)
(100)
(O)
(60)
(95)
(0)
THE PROBLEM
OF PRODUCING
(O)
(0)
(55)
32
groups to each other and to pauses which wore judged by
no ©bseFvere us limiting phrases#
©xe portions of speech
which lie between the phrase limits may be considered as
Ww e a l phrases*tf Portions in which both preceding and
following phrase limits are marked by 50 per cent or more
■of th© observers are compared objectively to similar portions
which were- marked by fewer of th® observer®,*
II.
EKSXJlfE
location and Humber of lhra.se Limits.
Figure I Illustrates that. In the first place, phrase limits
are not always located in the same places for the two per*,
formances*
For example, after the word "problem11 in the
second line from the bottom 95 per cent of th© observers
judged a phrase to end in the case of the speaking performance,
whereas no judgments were indicated for the reading of tlx©
speech.
$hls situation is reversed in th© cas© of the word
"producing11 in th© same line, where none of tlx© observers
judged a phrase to end in th© speaking performance, but 55
per cent of them indicated this point as limiting a phrase
in th© reading performance. 'Shat these differences are of
some magnitude is Indicated by th© fact that, of all th©
phras© limits selected by more than one-ha If of the ob­
servers, in all of the performances, only 56 per cent were
located at identical places In both speaking and reading*
A second difference Illustrated by Figure I is that there
33
^able IX*
Mean duy.atipn of
in seeomla*
Mean
H
Bar. S.P.
H
Ito. S.P.
77
*50
*58
40
.65
*56
117
.57
*
by
tsban 80
*06
*06
40
*05
*05
90
.06
.07
168
.08
.04
180
.07
.05
.08
.05
fudged by
50$ or
more
H
Pur. S.D,
34
&r© more phr&s© limit® In impromptu shaking than in oral
reading*
In th© ©as© of this one performance there are
per eent more phrase limit© in speaking than In reading.,
and when all of the performance® are considered, the excess
in favor of peaking is 59 per cent.
M&
of the Percenta&e of Judg-
ment© to tfos Duration of Pauses.
Table IX list® th© mean,
duration of pauses falling at phrase limits Judged by 50
per eent and more of the Judges, by fewer than 50 per cent
of the Judges, and pauses which were not indicated as
limiting phrases.
The number of pauses, their mean dura.*
tlon, and the standard deviations of duration values, are
included for each type of performance and for a composite
of both.
A striking difference is noted in Table IX be-
tween the mean duration of those pauses indicated most
Often and least often as limiting phrases*
¥hen considered
compositely, th© pause© most frequently judged a© phrase
limits are seen to be 0*57 seconds in length, or approxi­
mately ten time© as long as those judged less often.
As
might be predicted, these differences when compared for
©peaking^ reading, or for the composite distribution, are
reliable, as indicated by critical ratio® in excess of
three.
It appears clear that when the observers agreed
50 per cent or more of the time, they were indicating
phrase limits reliably different in duration from those
agreed upon by fewer observers*
Pauses not judged as
phrase limits are, in the case of both ©peaking and reading.
Table III*
Mean duration of phrae.ee in seconds*
Combined
JUdgad by
bo% or
more
Judged by
fewer than
Mean
Dur*
8. D*
Bur.
S.D*
Mean
Bur*
3.8*
1*44
*91
2.25
.80
1*73
1*03
*79
*es
1.11
.22
1.00
.55
Mean
3 6
0*02 seconds, or approximately one-third longer than pauses
marked by fewer than 50 par cent of the observers*
*Fh©
difference is reliable for reading, but not for speaking*
92his uneaspected difference in favor of pauses never chosen,
ail limiting phrases arises from the fact that, by definition,
every pause., whether judged as a phrase limit or not, is
represented by a period of silence on the objective record,
whereas some phrase limits were indicated by the observers
be occur at places where no objective pauses were made*
When the mean duration of pauses at phrase limits
in speaking is compared to that of similar pauses in read­
ing, the difference is minor and non© of such differences
were found to be reliable*
Darctl.cn of Phrase a ar$| the Humber of
Words For Phrase*
fable 111 presents the duration measures
for phrases which fall between phrase limits judged by 50
per cent or more of the observers, and phrase limits judged
by fewer than 50 per cent of the observers*
Observation
of this table will show that phrases of the former category
are longer in mean duration than those of the latter type*
All three of the differences shown in fable III are
statistically reliable*
It will also be seen that th©
phrases which are judged most often for speaking are defi­
nitely shorter than like phrases in reading*
Ihes© phrases
In speaking are 1*44 second© In mean duration a© compared
to a mean phrase length of 2*25 seconds In reading*
This
37
of
In word©
3«D»
Mean
S* D*
by
or
4*83
3*86
8.63
3.98
by
fewer than
2*99
1.97
4.62
2.66
more
38
Table V* Mean pitch ranjge of
in tones.
Per
cent
tHWHMpW •in* • •••
'. HMWr
" or
t.©crease
Mean Range of
Phrases Judged,
by 60% or more
4.03
6*05
33
Mean Hang© of
Phrases Judged
by Fewer Than 60%
3*14
4*61
32
39
clear out dlfference is reliable, as la indicated by a
Cf&tleal ratio of 5*3,
The same trend holds between the
phrase® in speaking and reading when the phrases judged
by fewer than BO per cent' of the observers are compared,
hat the critical ratio is only 2*1#
Table IV Is similar to Table 111* except that In
this ease# the measure is the number of words per phrase*
In Table IV it will be seen that there are definitely more
words In the phrase® judged by one-half and more of the
observers than there are in the phrases judged by fewer
than one-half of them#
These differences are reliable for
both speaking and reading#
Inspection of Table IV will
also show that the phrases most frequently judged in read­
ing contain definitely more words than do like phrases in
speaking#
This difference of 8*6 in comparison to 4*8
words per phrase for reading and speaking# respectively,
is likewise reliable#
That phrase® vary a great deal in
this respect can be seen by the large standard deviations#
4#
Pitch Character!sties of the Phrases#
Table
V state® th© mean pitch range of phrases In tones# th©
pitch rang© of a given phrase being th© difference between
the highest and the lowest points in that pox*tIon of th©
objective record#
Observation of the table Indicated that
the phrases judged more often have a greater mean range in
tones than phrases judged less often#
For speaking there
is a difference of 0*89 tones between these two types of
phrases# and for reading a difference of 1*44 tones in favor
40
phrases moat frequently fudged fey the observers.
In comparing th© average rang© a for Impromptu
©peaking and reading in Table V* it will fee seen that^there
1® a rang# of 4*03 and 6*05 tones for ©peaking and reading,,
respectively* or an increase in tones of one-half for read­
ing* when phrase a Judged by 50 per cent or more of the
observers are considered#
For the phrases judged less often,
a similar degree of increase is noted in the reading per­
formance*
It is ©f ^special interest to note that th© mean
pitch range of the phrase® In reading of 6*05 tones* or
approximately one octave* is a® great as that frequently
reported in textbooks and in earlier investigations for th©
total pitch range*
This may be explained in part fey the
fact that B<mm of the earlier frequency records were un­
readable for low frequencies.
In other cases* with ade­
quate recording method®* th© ©angles have not been suffi­
ciently longer than these phrases to give expression to
infrequently used high and low pitch values* which will
increase in number as the sample la increased in length*
As regard® th© effect of increased sample on pitch range*
it will be noted that th© phrases in reading are longer
than those in ©peaking and also greater in mean pitch range*
19a© phrases judged SO per cent and more of th© time are
longer than those Judged less than 50 per cent of th© time,
and again increased pitch range is seen in the type of
phrase representing th© longest sample#
41
Sable VI,
of hiKh-toefore-low and low-before'Bin phrases/ and dtreetion of
Pen m n t of Bl$bBefore-Low Points
65
Per ©out of Low-*
Befor©*-Htgh Points
47
73
Dominant Pitch
Movement
Pen cent Down*
ward Movement
41
©7
Per cent Up­
ward Movement
20
9
Level or
Amenable to
Istimatlon
eFor phrases judged 60$ or more th© time*
42
VII.
,
lea& extent of dominant intimations in tones
iSen»Of
MMfcfriSiii SPew*1
]^" pcenta
wii^>i^,iiiiibiii«
m m**
w
\
r ■»n111
■S|nwfriw ***• $iv©n
<1■
'
dlrocMon
of
M
t
e
h
movement*^
wiWMStwwiWin^M■■».«!>■*■***■.■■' ,
<i>iw»* jWlioi^i'i’Mm mu.
wmpuwri■■»>»»
Per­
centage
Extent
Percentage
Extent
Upward Intonations
Ending Phrases
5$
2* 12
55
5*08
downward Intonations
Ending Phrases
47
3*34
45
3.63
Upward Intonations
Beginning Phrases
65
1*72
75
2*62
Downward Intonations
Beginning Phrases
55
1*14
35
2*04
#For phrases judged SO pen cent or more of the time*
43
Table VI present® additional measures of pitch
in the phrases meat frequently Judged#
The- first measure
of "hii$i-b©fore-lowff and "low-before-high,# points merely
consider© which extreme pitch, either high or low* comes
first In the phrase*
The second measure, that of dominant
pitch movement* Is based upon a Judgment of the general
direction of pitch movement as seen from an inspection of
the objective record*
In the latter measure a number of
phrases were considered to be not amenable to Judgment.
Inspection of the table shows that speaking has almost an
equal number of hlgh-before-low and low~b©fore~high points
within the phrases*
In the reading performance the phrases
have definitely more higk-foefore-low points In the pro­
portion of 75 to 28*
From observation of the lower part
Of this table it may be aeon that in speaking 41 and 29 per
cent of the phrase a have downward and upward movement,
respectively.
In reading* 57 per cent of the xihrases are
dominantly downward in movement and 9 per cent upward*
In
each ease approximately one-third, of the phrases were either
level or not amenable to Judgment *
It appears clear that
the general direction in pitch movement tends to be downward^
in the case of both impromptu speaking and reading, but that
this tendency is more definite In the case of reading.
fable VII expresses the relative numbers and
extent® of the dominant pitch movements at the ends and
beginnings of phrases*
It la seen that approximately an
equal number of phrase® end with upward or downward intona-
J
J
nimber and mean extsnft jg£ shifts
fable VIII*
“ "t*
r"'“
”
w
Reading; of
Mean
Extent
%
Mean
Extent
Mean
Extent
Between
Phrase®*
Ikmnw&rd
54
2*3
2*5
26
1.7
Upward
34
3*1
4*6
74
4*6
bevel
12
0
5
Within
Hirases
Downward
20
1*5
18
1*7
27
1.6
Upward
53
1*3
58
1*6
48
1*9
bevel
87
25
05
eln the ease of this study, phrases falling between points
Judged by 50 per cent or more of the observers as limiting
phrases#
45
fcIona*
35i6 extent off such. pitch movement is greater in
phrases of the reading performance than in those off the
speech.
32ie extent off the downward movement at the end
Off phrases is greater in both performances than the extent
off such upward movement.
It will also toe seen from this
table that phrases in tooth performances begin more often
with upward pitch movement, and that this is ©specially
true off the reading performances*
When the mean extents
off sueh upward and downward movements are compared, it will
toe seen that upward pitch movements initiating phrases are
greater in extent in tooth speaking and reading.
Upward and
1
J
downward Intonations in goatling are greater in extent than
>
in speaking*
25b.® first two column® in Table VXXX present data
cm toe relative number and mean extent of shift® in Im­
promptu speaking and reading*
The third column presents
similar data from Pronevost^ on the oral reading of a
2*
Fronovost, W** /m experimental study of the habitual
and natural pitch levels of superior speakers* Ph.D.
Dissertation, State University of Iowa* 1939*
carefully written prose passage by these aam© speakers.
In­
spection of the vertical r@lationjsJ.iips in Table VXXI indi­
cates that for both reading and speaking* slightly more thanN\
i
one-half of the shift® between phrase® are downward in pitch ;
movement* somewhat more than one-third of them are upward,
and tfe* reminder are l w l t
Observation of tlx® horizontal
relationships in the first two columns of the table indioat©
that there are no striking diff©renews between speaking and
reading* but the extent of shifts la slightly greater for
^
both upward and downward shifts in the reading perfoamwui&©*/
In the ease of both performances* the upward shifts are
greater in mean extent than are the downward shifts*
Data
from the reading of the wall-written prose passage* presented
in the right hand column of fable VXIX shows a reversal of
this relationship insofar a® the relative numbers of shifts
between phrases are concerned* for in this type of perform-'
ane© upward shifts are definitively more common* 74 per cent
being of this type*
Agreement is seen* however* between the
two studies Insofar as the extent of such shifts is con­
cerned* for the upward tnterphrasal shifts are greater in
mean extent in all cases*
These discrepancies between the
findings of this study and those of Pronovost^* appear to
3* Fronovoat, W*, oj>, pit*
arise from the differences in the materials* because the
same six superior speakers were used in each study and per­
formed under like experimental conditions*
By way of comparison* shifts within phrases for
the three types of performances are seen in Table VIII to
be more often upward than downward*
The extent of such
*
upward shift a is slightly greater In Improis^tu speaking and
47
In the reading of written material than In the reading of
the speech*
In reading.*, the extent of downward shifts
within phrase® exceed® that of the upward shifts by only
0*10 tone* an amount too small to indicate that the trend
for this type of p®3?formanoe Is other than accidental*
in*
mrnum:
Recorded performances which served for a previous
objective comparison of impromptu speaking and oral reading
were employed as materials for an exploration of the char­
acteristics of ?lvocal phrases*M
Twenty trained observers
Identified the limits of what they judged to b© phrases as
they listened to these recordings*
Objective measurements
of the pitch and duration characteristics of phrases thus
Judged were made* with the following results#
1*
As the percentage of agreement in the
identification of phrase-limits Increased* the mean duration
of the pause® at the phrase limits beeam© progressively
longer*
B.
Pauses judged by 50 per cent or more of the
observers as limiting phrases were approximately one-half
second in length* and ten tlines as long as pauses judged
C® limiting phrase® by fewer than 50 per cent of the
observers*
5*
Phrase limit® were frequently judged to be
at different point® in the two performances*
48
4*.
Inhere were markedly more phrase limits in
the impromptu speaking performance than there were in the
reading performance*
5*
The pauses at phrase limits were longer in
reading than in impromptu speaking, but not reliably so*
6*
Phrases identified by 50 per cent or more of
the observers were reliably longer than those Judged by
fewer than on@~half of the observers*
7*
Phrase® identified by 50 per cent or more of
the observers were reliably longer in reading than in
speaking, but phrases identified by fewer observers were
not reliably different in the two performances*
8*.
There were more words per phrase In the read-*
ing than In the impromptu ©peaking performance*
9.
Pitch movements throughout entire phrases
tended to be downward* especially in reading*
10*
Phrases in both performances ended with an
approximately equal number of dominantly upward or down**
ward intonations? the downward intonations were greater in
mean extent*
11*
In both performances phrases began more often
with upward than downward Intonations# the upward intona­
tions being greater in mean extent*
12*
Shifts between phrases tended to be downward
more often than ^pwand, both in Impromptu speaking and in
oral reading of the speeches, and contrary to previous re­
ports on the Reading of written material*
Within phrases,.
49
however, the results agreed with other reports in showing
that upward shifts outnumber downward shifts*
The present
Study also confirmed previous data by showing that the
mean extent of shifts between phrases, is greater for upward
than for downward shifts*
m
xirmTiGATioii of m m ..pitch abb duhatioh qharao^bistxqs
OF THE ORAL REABIB0 PERF0RIIABCE3 OF
3BFBBX0R m m m speakers
^Experimental Investigations of the speaking voice
have been restricted almost entirely to male subjects*
With respect to pitch and duration It has been assumed
that the female voice, apart from the usual difference in
pitch level* is similar to that of the male, but the
similarities have not been demonstrated nor has the extent
of the difference in pitch level been quantified satis­
factorily*
I*
The work ©f Cowan3- might appear to require an
Cowan, M#, FltCh and intensity characteristics of
stage speech, Ar» Sg*, Suppl., 1936*
exception to this statement*
However, the nature of the
performances studied in that investigation do not allow
for direct comparisons between male and female voices, and,
furthermore, the results can in no sense be considex’ed as
describing the eharacteristics of the female voice during
the speaking or reading of
material.
Because
of this almost coa^Iete lack of Information, it has been
the purpose of the present investigation to describe
51
objectively toe pitch and duration character! aties of toe
Voices of superior women speaker® reading factual material,
and to compare these data to previously established values
for superior male speakers#
X*
MPKEimilTAh FEOQI^UCT
toe reading material in this Investigation is
toe same prose passage used by PMnovost^, C u r r y and
8« Pronovost, W#f to experimental study of the habitual
and natural pitch levels of superior speakers, Ph*D.
Dissertation, state University of Iowa, 1939*
3# Curry, E* T*, An objective study of the pitch ehar&©~
terlsties of the adolescent male voices, Ph*D* Disser­
tation, State University of Iowa, 1939*
McIntosh*4 The entire passage, as reproduced below, was
4* McIntosh, G# W* Jr», A study of toe relationship
between pitch level and pitch variability in the
voices of superior speakers, Fh*D* Dissertation,
State University of Iowa, 1959*
read by each subject, but only the italicised portion was
Objectively studied*
When the sunlight strike® raindrops In toe
air, they act like a prism and form a rainbow*
toe rainbow 1® a division of white light into
Sanv" ''beautjlfuX ecior®*''T"’
&ese""take TSie"sii'ape’'of
a long round arc^# w£th '&¥s"1
above,
an^
® W ® 'lappi3^nl
^Xy ^SsES^'Ss^SlSkSB^
fiirei aC^Ccor^nStoTe^nHTa^oTIXriX^^F^Df
Isoldat one end* People look* but no one ever
—
i $ m
a f f i m n b W T o r ' ^ o m e l b l H s TJeyond
M e reach* M i friends say lie Is. looking for the
pot of gold at toe end of the rainbow*
toe process of selecting the subjects parallels
tost ’
used in the above mentioned studies*
For purposes of
dOB^ariaon It was deemed essential that the women speakers
should evidence the same -superiority In general performance
toaraetwtsttos and in their use of voice as did the men
speakers*
to this end* 25 adult female subjects were
selected by the staff of the State University of Iowa for
their superior general effectiveness in speech and voice
usage*
toes® subjects were tested rigidly by the experi­
menter and the supervising staff member in a procedure
that included an articulation examination, an oral reading
performance, and an impromptu speech*
From this group*
eight subjects were selected who satisfied the requirements
for superior voice usage*
With the subject and the experimenter in a soundproof-room, phonograph recordings were made of four read­
ings of toe pros® passage for each subject*
In each ease
ample time was allowed for the subject to practice the
passage both silently and aloud*
to© following directions
were given before each reading*
Head toe passage as well as you are able*
as If you were reading to an audience of approxi­
mately 25* A phonograph recording will b® macte*
A five minute interval was 'allowed between each
of the four performances*
toe best of the four records
for each of the eight subjects was then selected by the
Table I*
Median
leyele for men and women*
Median
Io SSBl
level#
Women
MxSJjScts
Median
plilBS
TeW'X#
C
18*98
F
22*80
B
18*51
C
28*48
B
18*46
B
22*26
A
18*17
D
22*20
F
17*77
A
21*52
E
17.51
B
21*45
Composite
18*24
Men
i
#2n tones above aero freqaencv level*
82*19
method .of ranking # with ten trained observers serving as
judge©#
3&eh sub ject Vs best recording was retained Tor
further consideration and the other three recordings dis­
carded*
fhm
final .selective process consisted of rank­
ing the eight recordings by the matho-d of paired comparisons#
Fourteen trained observers compared each reading twice with
each other reading# one© before and one© followingj the
pairs were presented in random order#
On the basis of this
procedure# the six highest ranking recordings were selected
for further study#
«•
1*
m s m
Fitch Level#. Measures of central tendency,
in terms of median pitch levels# are presented in M b X © I
for the voices of the six women used in this study and for
the voices of the six men studied by Fronovost*5
M
.^■■IMI^.I.*I
|HM ■■■■■■■■■■II.......1.„
5#
.,r.,...;i|.
.,r.^
£h@
— 1—
-mr-*rvT"m-:rnir
Fronovost# w# # op* cit#
median pitch level® are stated in tones above
16#35
cycles
per second6 # and the values ranked in descending order*
6
|fl|^
#
$he zero reference frequency proposed by Fletcher, H# #
pitch.# and timbre of musical tones, j#a.s#a.,
6 # 1 9 3 4 # 5 9 -6 9 #
~ “---T.; II III III III-.IV^
—
j— —
«*' iM iritimiii—■nfrdMimmlil i m «.
y
■_ ...
,—
Inspection of the lower part of Tatla
f r - nm .. ■• | r y n
1
> ii'nin Hr■■n i mi r i ■"1' i~ r."i '[
'ri r
Indicates that when
55
$h© distribution© are considered ©ompositely# the median
pi toll levels of the women1© voices ere 4*0.5 tones higgler
than th© median pitch level® for men1s voices*
Mien tii©
median pitch level of each woman is compared to that of
@$|0h man# there .1® a mean difference of 3*89 tones between
the medians of the two sexes*
Individual differences of
some magnitude are Indicated by th© fact that the lowest
median pitch level for women Is only 2*53 ton©® higher than
the highest median level for men* whereas the highest
median for women is 5*29 tones higher than the lowest
median for men*
Critical ratios of a high order Indicate
that there Is a reliable difference between th© medians for
each on© of these 36 comparisons*
This finding of approxi-
mately four tones* or two-*thirds of an octave* difference
between the median pitch levels for the voices of woxnen
and men Is in contrast to th© general assumption of a on©
octave difference•
Hie highest female pitch level is less
than an octave higher than the lowest male pitch level.
When the difference© between median pitch levels
for each sex &r© compared in lable I* It will be noted that
they are small*
For th© male subjects* the difference
between the highest and lowest median pitch level is 1*41
tones; for the women# the difference is only 1*35 tones.
Although not conclusive proof# tills finding, suggests that
the preferred pitch lovely for th© .superior voices of both
women and men fall within relatively narrow limits*
Further-
more* there is a possibility that th© median pitch level
COMPOSITE
Figure I* >3uperimpoa ed epiiposite and Individual pitch
'
reaaln^ performance s
pj gen [and*vYOpey^s|^ak:6?sT Medians ape
l^nd^cute^h'y Kori iaontlai 'linea across each
<uLefcrib^ion
^TaS:
eTiecir fn cycles’per seconds
57
Of * given qpMke* may fro relatively independent of how
that vela* may b© classified in terms of its effective
eiuglmg ran®©*
In support of this po@aibll.ity it should
be noted that the female voices studied Included the usual
pitch categories of th© female singing voices of soprano,
second soprano.# contralto, and alto, and although the
hipest median pitch level for speaking was found for a
subject who stated that she sang soprano, the lowest was
revealed for a subject reporting that ©lie sang second
soprano*
Hie second* third* fourth* and fifth ranking
median pitch levels were found for subjects reporting them**
selves as singing second soprano* alto, soprano, and con**
trait©, respectively*
inspection of the literature on the
singing voice does not reveal optimum, or average pitch
levels for each of the part categories her© represented,
but it does appear clear that for these speakers the
preferred median pitch levels for speaking fall "within a
narrow rang© of tones regardless of the classification of
the individual singing voices«
Figure X is a graphic representation of the
distributions of pit die© used*
Hie horl ssontal lines rep­
resent the ©am© median pitch levels stated in Table X,
except that here they are given in cycles per second,
The superimposed distributions on th© left side of th©
figure are composites for all of the pitches used by th©
women and th© men.
From left to right, the other distri­
butions are those of th© women and th© men in order of
58
fablo XX#
;
* of total
m m iHWWmUn1
wmn
W&nffiB
Feroent&ge
ME®*
DQxOW
bo.Xow median,
wlow
28
B
A
A
20
C
20
B
17
D
23
E
25
B
21
E
G
9
Average
18
25
59
deccending .median pitch level*
Musical tones are plotted
along the ordinate, and th© percentage of total pitches
used along th© respective abscissae*
On th© left side of
pair of individual di strlbuttons# singling ranges are
represented by a solid**line- curve for the normal ranges
and by a dottsd^lln© curve for the falsetto register*
from an Inspection of Figure X it will be noted that the
distribution© of pitches used by the women and men speaker©
overlap to a considerable degree*
In order to arrive at
a Value representative of the amount of tills overlap, th©
highest pitch used by each male was Interpolated into th©
distribution of pitches for each female*
The average of
these 36 computations Indicates that 50 per cent of th©
pitches used by th© women fell below th© highest pitch©©
used by the men*
Individual amounts of overlap vary eon*-
sl&erably, however* 18 and 79' per cent being th© extreme
values*
Table XX expresses the fraction of th© total
singing range# including falsetto# which lies between the
median pitch and th© lowest sung ton© for both women and
men# th© mean for th© six women subject© being 0*18, and
that for the men# 0*85.
Fractions for each individual are
stated in order of decreasing magnitude*
In arriving at
those values# each subject sang down the seal© and up the
scale as far as It was possible to sustain a tone * From
the piano keyboard a notation was made in terms of the
nearest s©mi«*bom© reached in each of three trial© for the
fable III*
Sfcandagd ,rn
^ButTmaa
^tW»_
»»WW»»ll>»lli««IWWWWWi<wll
'<l»IIti:'■'
l4$sl$
x>
%t%r
2*14
E
M
2t
tonea used*#
Women
SuWjeets
A
1*81
1*91
B
1*89
A
1*88
0
1*68
B
1*88
F
1*54
F
1.81
B
1*40
C
1*70
D
1*57
D*
1*89
1* 56
^Arranged In order of decreasing standard deviations*
highest sustained tone# the lowest sustained ton®* and th©
last tone of the n o » l register,
bar five minute intervale*
The trials were separated
It may be noted from Table 11
that the median, pitch levels for women* considered composltely, are relatively lower in th© total singing range
than those of the men* and the fraction of their total
range which lies between the median and the lowest sung
tone is more variable than this same measure for the men#
A method of predicting the natural pitch on the
basis of this relationship has been shown by Fronovost^
7*
Pronovost* W*, ©g* olt*
to be one of th© most satisfactory clinical methods of
arriving at the natural pitch level of the ©peaking voice*
2he present data Indicate* however, that in applying this
clinical measure to women fs voices* the relative diff©rence
in the location of the natural pitch levels should be
taken into account, as well as the greater variability of
this measure#
2.
Variability of Pitch*
Table III presents
standard deviations of the pitches used by each subject In
th© order of decreasing values*
Observation of the lower
line In Table III Indicates that th© mean. S.D. of the
distributions of pitches for women is 1*56? tones* or approxi­
mately one-third of a ton© less than the mean 3*X>* of 1*89
fable Vfi
In tones of median
TSoHpoHTe
(Z/Jrxcou ( pifotiouasr)
Women
Men
Differea
di&n 10 per cent
w
H
20 «
♦58
.46
#08
.79
.92
.13
n
50
n
tf
1.25
1.45
.20
0
40
H
ti
1.70
1.97
.27
ti
50
tl
tf
2.18
8.52
.34
tf
60
tt
n
2.71
5.16
.46
«
70
«
«
5.56
3.82
#46
n
80
fl
H
4.07
4.92
.85
@0
n
ft
5.02
6.87
1*26
ft
10.46
10.52
.06
if
n
100***”
^Distributed equally above and below the median
**$Squlvalent to total pitch range*
63
t m e m tor th© distribution of pitches used by the males#
Wion individual difference® are considered, the standard
deviations of pitches used by five of the six women are
seen to be lower than all but one male value*
The S.B.
of 1*91 for Sub j©ct>A, female,. Is the same a® that for
the fifth ranking' s#D* for the male voices*
fable XV expresses the extent in tones of th©
median percentage® of pitches used*
lb© nature of this
computation as a measure of pitch variability has been
discussed at length above*
Briefly stated, the value for
the median 10 per cent is the range- in tones which subtend#
the pitches distributed between the 45th and the 55th
percentiles, etc#
Xt will be seen in fable XV that in all
cases the values are smaller for women than for men*
This
difference become© progressively greater for men for each
range up to and Including the median 90 per cent*
Th©
median 100 per cent, which is synonomous with th® total
pitch range, shows a difference of only 0*00 tones In favor
of tix© men*
This small difference in total rang© confirm®
th© suggestion by Lewi® and Tiffin® that the total pitch
8*
Lewis, D# and Tiff in, «T* , A psychophysical study of
individual difference® in speaking ability, Ar* Sg*
1, 1934, 13-80*
rang© doe© not adequately describe pitch variability*
«
If
values for the median 90 per cent ar© considered to describe
the Afunctional” pitch range, th© mean range of the female
64
H
> I
5 m
o
co
9 192
^ O S'?< 2 > </>5
q "H§ 11
m
Figure II#
*
Z m
1
Pitch curves of the voices of fggcle subject
ID (upper curveT ana E l e mitject E (lower
curve) reading Identical proa© material*
65
subject® £« Mon to be 1*25 tones narrower than that of
the comparable male speakers, a difference amounting to
an Increase of 25 per cent In favor of the latter
box*
Among the individual subjects the widest median 90 per
cant range of the females is exceeded toy five of the six
male subjects#
It is difficult to explain why the voices of
superior women speaker® are less variable in pitch than
the voices of superior men speakers*
However, previous
data® indicate that low pitched, performances are relative9#
Xy
McIntosh, C. W# Jr* , og# eit#
lacking In pitch flexibility, and it will be recalled
that the speaking voice® of these female subjects are
relatively lower pitdied in their total singing ranges than
those of the male subjects#
$*
^honfttlons# Inflections., and Shifts# Pitch
curves of female subject I? and male subject E are shown In
F ig u r e
xi for the first four seconds of their respective
performances#
From such pitch curves of each subject the
following measurements were Bade:
(1) The extent and
duration of phonationo, (2) the extent, duration, and
direction of inflections, and (3) th© extent of shifts,
both within and between phrases. 10 With the exception of
66
fable V*
Mean extent of phonatlpna* infXeetlens* and
Women
lien
lean Extent of Phonatlons
2.6
5.2
Mean Extent of Inflaetlems
All
Downward
Upward
B#1
2*1
2.1
2*3
2*4
2*1
Mean Extent of Shifta
H t W n Phrases
All
-
*6
*2
*5
* .0
1*6
1*8
*2
Downward
1*6
1*6
*0
Upward
1*7
1*9
*2
Mean Extent of Shifts
Between Phrases
Downward
Upward
^
2*3
1*Q
2.9
3*5
1*7
4*6
1.2
***1
1*7
I®*
Phram ® to tilts study were determined In the following
manner: Nausea which were used by at least ore~half
of the speakers and which were longer than adjacent
pauses were selected arbitrarily as totei^phrasal
pauses.
Shifts between phrases, these measures hare been defined
above.11
11* Fairbanks^
, Eecent experimental Inv©stigations of
vocal pitch in speech, £.A.S,jV* 11, 1940, 457~466.
Inspection of Figure II indicates* (1 )
a remarks
able similarity to th© pitch and duration pattern of the
two subjects, (2 ) the tendency of the woman*s voice to
have most of it a pitches distributed approximately four
tones
above those of tlx® man., and (5 ) greater flexibility
to the male vole® than to the female voice, as Illustrated
by somewhat greater mean extents of pixonations and In­
flections.
Table V lists the mean extents to tones of
donations. Inflections and shifts*
It will be noted that
men exceed slightly, or are equal to, women to all of these
mmmzp®®, except that of downward shifts between phrases
in which there is a difference of only 0*10 tone® in favor
of women*
to general, the differences to favor of men
m m not great, but their consistency Indicates that to
respect to these measures also'the women*s voices are
68
Table VI*
ends
Shifts Between Phrases
Upward
57.14
Downward
26*19
Level
16.67
Intonations lading
a Phrase
Upward
SI. 25
Downward
64*60
Level
4*20
Intonations Beginning
a Phrase
Upward
Downward
Level
*&«SB
sa.io
9.52
74
0
69
variable than those of the M i
¥©riia*a inspection of the values in Table ¥
indicates that women and men are much alike in respect to
the#e measures.
In both cases the extent of upward shifts,
within and between phrases,, tends to be greater than the
extent of downward shift®*
The mean extent of Inflections
for women is the same for both upward and downward inflec­
tions.
For men the downward Inflection® are one-third of
a tone greater in mean extent than are the upward inflections*
Table ¥X present® data on pitch characteristics
at the end and beginning® of phrases*
Begar&ing shifts,
inspection of Table VI shows^that both men and women have
more upward than downward shift® between phrases.
It will
also be observed that men exceed the women in the number
of such upward shifts*
’
Bio percentage of downward shift®
Is the same for both men and women, and approximately onefifth of the phrases for the women*© performances are
level, whereas the men have no level shift©.
Regarding
intonations, the data for the women indicate that they end
phrases more often with a downward pitch movement, and
begin phrases more frequently with an upward pitch movement#
In impromptu speaking and reading, the men were seen to end
phrases about equally often with upward or downward pitch
movement, but to begin phrases more often with, upward
pitch movement*
This suggests that phrases in different
type© of performances may characteristically begin with u.p~
ward pitch movement, but that the maimer in which phrases
70
fakl©
at a i s st
*s»
{Qompo sit©)
sbaiss^
Diff* (Men
over women)
women
Hearn Bat© of
Fiteh Chang©
2.72
9*5$
12*27
Ml
15*4$
17.66
2*23
Upward
15.97
18.47
2*50
Downward
14.
17.00
2*02
Mean Bat© of
Fitch Chang©
During Inflections
Mean Bo* of Changes
of Fitch Per Second
During Phonation Only
All extents
6.2
7*4
1*2
Semi«*t<m© or
greater
4.6
4*3
*0
Leas than a
semi-tone
1.9
3*1
1*2
All extents
15.6
6*6
1.0
Send^tone or
greater
4.1
4*3
.2
l & m than a
semi-ton©
1*5
2.3
*8
Including Shifts
#In tones per second*
pnd with respect to pi,toll lavement may vary from on© type
Of performance to another,
*«
JlS&g S£
and aftxfflwr of Qtmimm*
fable VII pro sent a data fox* the rat© of pitch change, and
the number of change© In direction of pitch per second-,*
measure® which h a w been defined above*
It will be seen
that wessonfs voice© have In all case© a slower mean rat©
of pitch change than have the men*© voices*
This differ-
one© in the rate of pitch change may be related to the
fact that women1s vole©® appear to be less flexible, a©
Indicated by th© measures of variability previously dls-Cusaed*
IS*
In tills connection liclntosh.^-s found that rat©
McIntosh, G* w* Ir*, ©$>* pit*
of pitch change was slower in less flexible readings than
In normal readings*
Inspection of the lower part of Male VXI shows
that the mean number of changes In the direction of pitch
per second are fewer for women than for men when all extents
and extents of less than a semi-ton© ore considered#
She
mean number of changes are the same for changes of a semi­
tone or greater*
Mien the mean number of changes* including
shifts* are considered, women show fewer changes In direc­
tion for the three categories of change.
However, these
differences in the mean number of changes in the direction
7JL
Table VIII,
m m
of rat®.
(CoEtposite)
1$0wm
$o$*ds ? w Itate
161*5
Men
105*0
Total Heading Tima
20*45
19*90
Total Flionatod Tima
12*93
14*84
Mean K&tio of Phonated
Time to Total Time
*?3
/A
of* pit oh may be explained# in part* by toe fact that the
mon road the same pa®sag© aomewhmt more rapidly and there-*
f ore would' normally hair® more changes in the direction of
pitch per unit of time*
6*
m »& Measure a of Bate*
Table ¥111 state®
,
toe results of gross time measures for the performance®
of the men and women subjects*
Observation of this table
Indicates that the women read at a mean rate of ,161*5
word© per minute, a value which is very similar to that
of the men*
When these values are compared with norms
reported by Barley^ * however#, which are based on the
IS*
Barley# F* L* # A normative study of oral reading rate#
M.A* Thesis# State University of Iowa# in progress*
oral reading perfomaaces of a random sample of college
freshmen, the rate for women falls between the 55th and toe
40th percentile# and to© rate for men between toe 45to and
toe 50th percentile *
The third Item in this table shows
that toe total phonated time 1® relatively longer for men#
with a ratio of total pJaonated time to total reading time
of 0*75 for men and 0*63 for women*
This value gives a
measure of how much relative time is spent in phonation as
compared to pause time in each performance*
In general#
voices with .low ratios of total phonated time to total
reading time will tend to be judged as more staccato, than
voices with M.$x ratios*
fable IX*
Mcan duration at obon&tlons*
f^cn
Mean Durations of*
Phonations
*40
M X pauses
*84
Pauses within phrases
*10
Pauses between phrases-
*75
All inflections
*16
DmOTSird inflections
*16
Upward Inflections
•17
•&In seconds*
M l Inflections
74*0
100
87*0
100
Bownward Inflections
58*2
52
46*8
54
Upward Inflection®
31*2
42
38*2
44
tCt C
Level Inflections
4*7
0
2*0
2
M l -Shifts Within
Phrases
25# 0
100
22*8
100
0.2
37
6*2
27
10*8
43
11*0
48
5*0
BO
5*7
25
Downward Shifts
Upward Shifts
Level Shifts
*&3hiffca between phrases are, by definition, equal in number
from reading to reading and are not considered in thls t able-
jyy*gB*
Table IX presents the mean duration of pten&fclons,
pauses, and Inflections *
Observation of this table show a
that the moan duration of phon&tion© la much shorter In
the performances of the woman subjects than in those of the
m m # or In other words, the women1a voice a s&ght be char*
acterl&ed as being more “staccato* w
This Is the only
important difference revealed by this series of measures#
The mean duration ©f Inflections la seen to be slightly
longer for women, the mean duration of all pauses is very
slightly longer for men, the length of pauses within
phrases Is identical, and interphrasal pauses are longer
by only 0*06 second for women*
**♦
humber of Inflections and Shifts#
Table X
lists the mean numbers and percentages of inflections and
shifts*
Inspection of this table indicates that for the
55 Word passage women have 74 Inflections as compared to
87 for men*
There are more than twice as many level
Inflections for the women than for the men*
Downward in­
flections are more numerous than upward inflection© for
both the men and the women, and In about the same proportion*
Both sexes have more upward than downward shifts#
have more shifts within phrases#
Women
Level and downward shifts
are In about the same proportion for each sex#
The ff©3ffl©gatie;ffo of Oases Within the Interval.
fj*om the Lowest 3ppken Tone, to the Lowest Sustained S\mi%
77
Table XI,
Interval from
0
B
D
A
lowest Ton© of
Singing Bang© #
F
B
20*50
19*50
19*50
18*50
19*00
18.00
lowest Spoken
Ton# &
14*15
14*08
15*09
16*51
18*40
18. 79
6*55
8*42
4.41
8*19
#60
-*79
•06
*05
*04
*04
*02
*00
Bifferenee 2
Is below 1
(ton##)
Percentage of
Oases
Male Su
D
B
0
A
E
F
Lowest Tone of
Singing Range &
16*00
16*00
16*00
14*00
12*00
12*00
Lowest Spoken
Tone *
11*38
12*98
10*19
15.45
9*97
11*68
Difference 2
1# Below 1
(tone#)
4.12
5*02
5.81
*85
2*08
.38
*14
*11
*06
*03
.03
•00
Percentage of
Oases
*In tones above aero frequency level.
78
SSS&L*
tabulations in m m
section or Table XX repr©~
sent, in order, the lowest note or the singing range, the
lowest spoken pit oh, the number or tones between these two
pi tehee, and the percentage & of oases falling within this
Interval#
The eoiuputatIon s in Table XX were made became©
of an interesting phenomenon observed in the objective
pitch 'records#
Hot 1nfraqu©ntly long downward inflections
were noted, which fell below the lowest note of the singing
range*
Ihen the number of pitches falling between the
lowest spoken tone and the lowest sustained sung tone are
considered as a fraction of the total number of pitches,
the mean for the women is 0*04, and that for the men,
0*06*
It will be noted In the thii^d Item of the table that
these pitches are distributed in a rang© of from 0*60 to
6*35 tones for the various female subjects and from 0*38
to 4*12 tones for the male subject a*
Subject B, female,
Is seen to have no pitches distributed below the lowest
sung tone*
It Is suggested that these tones which fall
below the lowest sustainable sung tone may result from
relaxation of the vocal cords as phonatlons are ended*
Most of these very low pitches are to be found at the ends
Of long downward inflections which terminate phonationa*
Examples may be m e n In Figure II*
Six carefully chosen superior women speakers read*
a 56 word pros© passage under controlled experimental
conditions, and a phonophotographlc technique was used to
make a physical analysis of the pitch and duration charac­
ter! sties of the performances.
The data were compared to
those previously reported for superior male speakers read­
ing the same material, with the following results:
1.
fitch levels for the women1© voices were found
to be placed approximately two**thirds of an octave above
the pitch levels of the men1© voices*
2*
The median pitch levels for each group of
subjects were located within limits of less than one and
one-half tones, which suggests that the preferred pitch
levels for superior speakers may fall within relatively
narrow limits*
3*
The superimposed pitch distributions were
found to overlap to the extent that, on the averag% 50
per cent of the pitches used by the women were lower than
the hipest pitches used by the men*
4*
That fraction of the total singing range
which lies between the median and the lowest sung tone is
0*18 for the women in comparison to 0*25 for raen*
3*
The voices of the women were, in general,
found to be less variable !rx pitch than those of men, and
this relative lack of variability may be a function of the
80
lower median pi tab. level© of the women*
Total
pitab range a were approximately the ® m m in both group a*
However* the Afunctional11 pitch range® of the men were
found to be greater#
V,
6*
I
.
The mean extent® of phonations# Inflections,
and shift© were found to be ©lightly lose for the women
than for the men speakers, with the exception of downward
shifts between phrase®* in which the women slightly exceeded
the men*
7m
In both set® of performances upward shift®
exceed downward shifts* both in number and in mean extent.
Downward Inflect Ions were more numerous than upward in­
flection® and were not markedly different In mean extent*
8*
Both sexes used more upward than downward
shifts between phrases* and the upward shifts exceeded
the downward in mean extent*
In the performances of the
women* phrases ended more frequently with downward in­
tonation®, the mean extent of such movements exceeding
that of the upward movements; they began more often with
upward Intonation®* such intonations being greater in mean
extent than the downward Intonations#
Comparable value®
were not available for the male subjects#
9#
When compared to the men1s voices* the
women's voices were found to have & lea® rapid mean rate
of pitch change# and fewer changes in direction of pitch
per unit of time*
81
10«
Both men end women need slightly slower-
oral reading rates than that of the median subject in a
random ©ample of college freshmen*
In this respect the
difference between the two sex group® was small*
^ 11*
Women1s voices were found to have notably
s h o r t e r mean duration of phonationai the ratio of phonated
time to total time was lower than in the ease of the men1&
voice®*
> 12*
^h© speaking voices of both men and women
©meoas&jass pitches that fall below the lowest ton© capable
o f being sustained by the singing voice*
Such pitches
were found to compos© approximately five per cent of the
total pitches used in the performances#
APPENDIX tO PARES I AE© II
Appendix A
SABI® I.
Median Pitch Levels of Individual Subject a.
Subjects
Speaking
Reading
A
16.79
17*60
©
XT* 58
18.22
C
18.96
18.98
©
16.83
17*90
E
16# TO
17.48
F
16.2?
16.71
Comp*
17.30
18.00
* tones above 16.55 cycles per second.
SABI® II.
Median Pitch Levels of Individual Subjects.
Subjects
Speaking
ReadJ
A
113
124
B
124
133
C
145
146
D
114
128
E
111
123
F
105
111
Comp.
120
129
» Cycles per second.
Appendix A
III.
Mean Pitch Levels of Individual SubJe
Subjects
Speaking
Beading
A
16.76
17.58
B
17.61
18*22
G
18.98
19.13
0
16.91
17.88
B
16.69
17*46
F
15*99
16.70
Comp.
17.30
13.19
* Tones above 16*35 cycle® per second*
TABLE IV.
Mean Fltofo Levels of Individual Subject®
Subjects
Speaking
Bead:
A
113
124
B
124
134
C
146
148
D
115
128
E
111
123
F
103
111
Comp.
120
132
* Cycles per second*
84
Appeft&tX B
TABLE I.
Mgdlan Pttoh fcOYaXa M
Total gltpfe Ranges
Individual 3ub.1eots.
Speaking
Subjecta
Median *
Range **
A
U3
8.87
B
184
6.19
0
145
6 .3 0
D
114
7.87
E
111
12.81
V
105
8.55
Comp,
120
8.42
Reading
Subjects
Median *
A
124
6.61
B
133
9.17
C
146
11.68
D
128
3.65
E
123
14,48
F
111
9.07
Comp.
129
9*89
* Cycles per second.
** Tones.
Range **
85
Appendix C
fABLE I.
Standard Deviations? of Individual Bie^rlbutiOM*
Subjecta
Speaking
Heading
A
1*18
1.51
a
1,24
1.66
c
I# 21
1.47
0
1*15
1*71
E
1*63
1*99
F
1*49
1*53
1*32
1.65
Mean S*D*
86
Appendix 0
c*p«3# coryegpoafttgg to Various
M
s
s
Bgar
Subjects
A
B
C
0
12*
F
Comp.
looje
40.08
42.91
43.63
41.46
42.91
40.18
43.63
n%
37.57
39.56
41.86
37.73
37.99
36.25
39.87
90%
36,41
38.51
41.18
36.85
37.26
35.59
38.56
85%
35.0©
37*97
40,64
36.20
36,48
35.04
37.71
80%
35.37
57.43
40.14
35.66
35.81
34.65
37.07
75%
34*96
36.94
39.71
35.18
35.33
34.25
36.59
70%
34*66
36.63
39.30
34.81
34.88
33.85
36.13
65%
34.36
36.32
38.92
34.51
34.50
33.45
35.71
60%
34.06
36,01
38.59
34.21
34.12
33.04
35.32
55%
33.81
35.59
38,25
33.92
33.76
32.69
34.96
50%
33.58
35.17
37.92
33.65
33.41
32.34
34,59
m
33.35
34.78
37.59
33.39
33.06
32.00
34.25
hO%
33.12
34.42
37.26
33.13
32,68
31.53
33.89
55%
32.81
34.06
36.73
32.85
32.29
31.05
33.51
50%
32.43
33.69
36.57
32.55
31.91
29.57
33.13
*5%
32.06
33,33
36.22
32.28
31.57
29.99
32.67
ZO%
31.61
32.96
35.84
32.00
31.23
29.41
32.19
15%
31,14
32.45
35.41
31.59
30.81
28.80
31.64
10%
30,57
31.92
34.96
31.18
30.27
28.15
31.06
5%
29.86
31.13
34.20
30.46
29.18
27.84
30.81
87
Table I (eont.)
SubJaotB
A
3
*a i
P
XU
F
"w»»
44*^4
<■
4 5 .1 7
40.38
46.87
39.81
42.39
43.21
41.55
41.23
38.81
41.81
38.99
41.08
42.22
40,28
39.69
37.32
40.49
•*''T3r 39.21
38.66
36.45
39.54
38. 3S
37.80
39.36
40.79
38*49
37.88
35.86
38.78
37.28
38.62
40.27
38,05
37.36
35.47
38.11
36.78
37.90
37.56
36.88
35.07
37.64
36.31
37.49
39.27
37.21? 36.47
34.64
37.21
35.89
37.11
38.81
36.79
36.07
34.19
36.79
35.55
36.75
38.39
36.29
35.52
33.79
36.39
35.20
36.44
37.97
35.69
34.97
33.41
35.99
34.85
36.12 37*64 35.32
34.62
33.04
35.55
34.48
35.77
37.32
34.85
34.27
32.59
35.11
34.11
35.39
36.99
34.42
33.91
32.13
34.66
33.67
35.02
33.98
33.40
31.67
34,20
36.30
33.43
32.77
31.20
33.66
33.20
15%
0
C
32.70
33.67
35.92
32.36
31.81
30.72
33.05
31.98
32.87
35.40
32.26
31.26
30.23
32.27
31.07
32,14
35.79
31.24
30.38
29.64
31.39
31.10
34.88
30.27
28.77
23,84
30,29
88
Appendix &
•SAMIM I,
im&b of Median
JaEfeenia
Subjects
A
.83
•48
.78
i.ia
1*45
.Median
Mediae
.Median
Median
Median
Median
Median
Median
.Median
Median
1 .8 8
**
Median 3L0?g
Median 20£
Median 20$
Mad Ian k0%
Median
Median &
Median
Median U
Median 90%
Median 100% **
2.33
2.98
1 *Jf
8.87
B
0
.41
.33
.67
.8 0
1.13
.1*10
1.47 1.37
1.81 1.75
2.24 2.15
2.78 .2 .6 2
3.30 3.11
4.22 3.83
6.19 6.30
A
B
.35
.32
.67
1 .0 5
.36
*75
1.14
1.44
1 .5 6
2.85
3.59
4*47
5.65
9.17
1.99
2.44
3*03
3.22
4,17
11.68
1 .1 0
1.56
2.04
2.55
3.19
3*46
4.46
6 .6 1
2 .1 2
c
P
P
.2 7
.,35
.54
.72
.83 1 .1 1
1.13 1.49
1.45 1 .8 8
1.83 2.29
2.31 2.34
2.84 3.50
3.67 4.41
7*87 12.81
0
.50
.87
1.40
1.79
2.31
2 .8 2
S
.45
.90
1 .2 8
1.74
2 .3 0
3*04
3*48 3.70
4.52 4.71
5.64 6.23
8.65 14.48
*35
.32
1 .2 0
2.14
2.38
2.62
Comp
.31
.72
1 .1 0
1 .5 0
4.21
8.55
1.96
2.44
3*04
3.75
4.53
13.17
P
Comp
.38
.42
*34
1.23
1.72
3 .1 2
3 .7 2
.8 0
1 .2 6
1.70
2.14 2 .2 3
2.57 2.87
3.11 3*64
3.84 4.55
4.99 5.76
9.07 15.03
* Distributed equally above and below the median.
** Equivalent to total pitch range.
89
Appeaaix F
TABLE I»
Mgan
of PfrpnaUona. Infloot long and
SubJ«ot«
A
Mean Extent of
Fhonations
B
C
%
B
2.68
2.48
1.94
1.42
2.75
1.6A
!.«
1.47
1.51
1.7®
X.?®
InflootIons
1.71
1.95.
1.46
1.34
1.63
1.85
Upward
Inf lections
1.5®
1.49
1.51
1.77
1.88
1.7©
All
Shifts
2.14
1.91
1.65
1.81
4.32
3.66
Downward
Shifts
1.79
2.06
.99
1.62
1.89
3*61
Upward
Shifts
2.35
1.40
2.16
1.91
6.59
3.90
All
Shifts
1.75
1.42
1.62
1 .5 2
1.70
1.50
Downward
Shifts
1.92
1.01
1.21
1.54
2.29
1.05
Upward
Shifts
1.10
1.95
1.76
1.51
1.43
1.71
Moan Ext ent of
Inflections
All
Inflections
Downward
Moan Extent of
Shifts Between Phrases *
Moan Extent of
Shifts within Phrases
* fudged hy §0 per cent and more of the observers*
90
Appendix F
TABUS X (cent.)
Reading
Subject#
A
Kean Extent of
Phonatlona
B
G
D
E
F
2.34
2.98
1.86
2.79
3.04
2.29
All
Inflootlone
1*96
1.83
1.76
2.09
2.00
1*^9
Downward
Inflootions
1.98
1.99
1.83
2.13
2.09
1.61
Upward
Inflections
1.98
1.70
1.71
2.03
1.95
1.41
All
Shifts
2.49
2.79
2.93
3.25
5.21
3.56
Downward
Shifts
2*01
2.27
2*42
2*56
2*80
3.20
Upward
Shifts
3*4?
4.10
3.44
4.17
6.81
4.45
1.55
2.05
1.49
1.71
2.63
1.4§
.87
1.27
1.26
1.89
1.58
1.19
1.80
2*63
1.54
1.38
3.11
1.99
Mean Extent of
Inflections
Kean Extent of
Shifts Between Phrases •
Kean Extent of
shifts Within phrases
All
shifts
Downward
Shifts
Upward
Shifts
* Judged by 50 per cent and acre of the observers.
91
Appendix 0
*ABLS *• M a m m n 21
change and lumber si
Subjects
A
lean Hate or
Pitch Change *
B
0
C
B
P
7.38
7.23
9*18
7.57
8.81
8.29
All
Inflections
1 2 .3 0
14.63
12*56
14.05
14.43
12.57
Downward
Inflections
12.79
14.36
10.16
12.34
1 3.66
12.75
Upward
Inflections
1 1 .6 8
16.21
16*15
17.78
15.55
12.40
Moan Hat© of
Pitch Change
haring Inflect Iona
Mean Number of
Changes in Di~
rootion of Pitch
per seconds
During Phonation
Only
All Extents
6.69
7 .2 2
7.44
7.39
7.11
6.27
Semi-ton©
or Creator
4.46
4.98
4.06
3.76
4.60
3 .2 0
Dess than
on©
2.23
2.24
3.38
3*64
2.51
3.07
Including
Shifts
All E x te n ta
6.29
5 .8 2
8.06
7.30
6 .7 8
6.25
S e m i-to n e
or Ur e a t e r
4.37
4.08
5.50
4.21
4.76
4.00
1.93
1.75
2 .5 6
3.10
2 .0 2
2 .2 6
Less than
S e m i-ton©
* Tones per a©©end*
Appanaix a
SABUE X (oont.)
&M&4SB
Subjecta
B
0
D
E
W
9*76
12.07
9 .8 6
10*96
1 2 .9 8
10*06
All
Inflections
12 #64
18.64
19.63
1 8 .1 0
13.03
15.08
Downward
Inflections
12*66
19.19
19.33
13.34
16.71
14.83
Upward
Inflections
12.73
18.11
19.71
25.17
2 0 .3 0
16.60
5 .8 8
8 .0 1
6 .9 2
7 .2 0
7*18
7.57
Semi-ton©
or Greater
3.96
5.73
4.05
4.67
5.05
4. 86
less than
Semi-tone
1 .9 2
2 .2 8
2 .6 8
2.53
2 .1 2
2 .7 2
6.29
7.45
7.94
6 *9 2
7.03
7.03
or Greater
4.72
3.43
5 .2 1
4.84
5 *2 2
5.18
Less than
Semi-'tone
1*57
2*02
2.73
2.08
1.81
1.85
A
M©an Rat© of
Fitch Change *
Moan Hate of
Fitch Change
Buying Inflections
Mean Mtnaher of
Changes in Di­
rection of Fitch
per seconds
During Fhonation
Only
All
Intents
Including Shifts
All Extents
Semi-tone
* Tones per second#
AppQHdiX H
tabus x.
M t m m
iubJeots
si Sa&£ iss.
SBjftM as
Subjects
A
Words per
Minute
B
c
D
£
P
136,66 147.17 173.58 173.40 157.02 132.49
Total
Tine *
36,86
24.04
24,19
24,23
23.31
25*27
Phobated
Time *
25.U
14,27
16.00
16.90
16.73
14.99
.6 8
.59
,66
.6 8
.72
.59
11.73
9.77
9.19
7.73
6 .5 8
10,31
Mean Duration
ef PhonatIons *
.52
.46
.27
.34
.38
.32
Kean Duration
ef Pauses *
.25
.33
*14
.16
.17
*22
Mean Retie of
Phonated Tlse
to Total flee
Pause
Time *
Number of
« Seoonds.
47
30
58
47
39
46
94
TABUS I (cent*)
finite
Subloot*
A
B
C
D
E
F
w w & s per
Minute
203*72 191*53 173.58 181.62 181.92 16 2 .7 2
Total
Time *
24*15
17.85
24.19
2 5 .1 2
19.92
19.50
Fhomated
Time *
17# 19
12.73
17.04
15.42
13.65
12.15
*71
.71
.70
.67
•69
.62
7*00
5.15
7*15
7*70
6.27
7.38
Mean Duration
of Phonations *
*37
.42
.2 8
.38
.39
*31
Mean Duration
of Pauses *
*15
.18
.1 2
.19
.1 8
.2 1
Mean Ratio of
Fhonated Time
to Total Time
Pause
Time *
Pauses *
* Seconds*
46
29
59
40
35
38
95
Appandta i
I HAVE ALSO (65) HAD A LONGING (35) TO TEACH (60)
DRAMATIC LITERATURE (95) ONE (75) OF THE THINGS I DID (15)
IN ORDER TO PREPARE UK SELF FOR MX JOB (45) WAS TO GO TO A
SUMMER STOCK CCMPARE (30) FOR TWO SUMMERS (100) WHILE I (15)
DURING (65) AND (10) FOLLOWING (15) THE TEARS (55) WHICH I
TAUGHT (100) AND (60) THEN (75) WHEN I (45) FOUND I WAS
ABLE TO (70) I CAME HERS (10) TOT H S UNIVERSITY OF IOWA (95)
AND (5) TOOK UP (60) WORK (70) IN THE THEATRE PROPER (100)
I (75) HAVE (70) MADE AN ATTEMPT (9©) TO (65) CONCENTRATE
ALL MX ACTIVITIES (75) UPON DIRECTION (50) AND ACTING
Raadlm
I HAVE ALSO (5) HAD A LONGING (25) TO TEACH
DRAMATIC LITERATURE (100) ONE OF THE THINGS I DID (15) IN
ORDER TO PREPARE MXSELF FOR MX JOB (100) WAS TO GO TO A
SUMMER STOCK COMPANY (15) FOR TWO SUMMERS (95) DURING AND
FOLLOWING (10) THE YEARS WHICH I TAUGHT (100) AND THEN (10)
WHEN I FOUND I WAS ABLE TO COME HERE (5) TO THE UNIVERSITY
OF IOWA (100) AND TOOK UP MX WORK IN THE THEATER PROPER
(100) I HAVE MADE AN ATTEMPT TO CONCENTRATE ALL MX ACTIVITIES
(5) UPON DIRECTION AND ACTING
96
Appendix 1
tabu:i
(oont.)
Subject B
I WANT TO GO OH THK STAGS (10) EVENTUALLY (90)
AND 30 (15) I BELIEVE (5) THAT THE BEST (05) PREPARATION
FOR THAT (05) IS TO BE OH THE STAGE (15) AS W C H AS
POSSIBLE HOW (9 5 ) HERE AT THE UNIVERSITY OF IOWA (95) I
GST AN OPPORTUNITY (85) TO APPEAR ON TIE STAGE (50)
BEFORE AN AUDIENCE (95) IN GOOD PLAYS (95) PLAYS (60) OF
CONTEMPORARY LIFE (95) AND PLAYS OF HISTORY
twnnya
I WANT TO GO ON THE STAGE (20) EVENTUALIY (100)
AND SO (15) I BELIEVE THE BEST PREPARATION FOR THAT (95)
IS TO BE ON THE STAGE A3 MUCH A3 POSSIBLE (30) NOW (90)
HERE (5) AT THE UNIVERSITY OF IOWA (95) I GET AN OPPOR­
TUNITY TO APPEAR ON THE STAGE (10) BEFORE AN AUDIENCE (100)
IB GOOD PLAYS (90) PLAYS (10) OF CONTEMPORARY LIFE (95)
AfiD PLAYS OF HISTORY
m
Appendix I
TABUS J (oont.)
Sublet e
AND THAT OP COURSE (10) COMES THROUGH PRACTICE
(10) IN MEETING PEOPLE (100) THE SECOND THING (10) I
THINK (SO) WHICH IS MORE IMPORTANT (10) IN PREPARING FOR
A FUTURE JOB (100) IS THAT (15) EACH ONE OF US (5) MUST
BECOME ACQUAINTED (10) WITH (15) THE FUNDAMENTALS (100)
UPON WHICH THE BUSINESS ORGANIZATION OF THIS COUNTRX WORKS
(100) THAT OF COURSE INVOLVES (5) TAKING POLITICAL SCIENCE
(15) AND ECONOMICS (100) AND STUDYING (85) VEffit CAREFULIX
(95) THE PRINCIPLES UPON WHICH (5) OUR BUSINESS ORGANIZATION
IN AMERICA (10) IS BASED
A ® THAT OF COURSE (35) COMES THROUGH PRACTICE
IN MEETING PEOPLE (100) THE SECOND THING I THINK (85)
WHICH IS MOST IMPORTANT (15) IN PREPARING FOR A FUTURE JOB
(100) IS THAT EACH ONE OF US MUST BECOME ACQUAINTED WITH
THE FUNDAMENTALS (85) UPON WHICH THE BUSINESS ORGANIZATION
OF THIS COUNTRX WORKS (100) THAT OF COURSE (10) INVOLVES
TAKING POLITICAL SCIENCE (30) A ® ECONOMICS (100) AND
STUDYING VERX CAREFULIX THE PRINCIPLES UPON WHICH OUR
BUSINESS ORGANIZATION IN AMERICA (55) IS B A S ®
98
Appendix I
TABUS I (oont.)
Subject D
I THIHK THERE ABB ABOUT FIVE POINTS (60) IB (?0 )
BSC BACKGROUND ABU TRAINING (20) THAT I MIGHT MENTION AT
THIS TIME (100) ONE OF THEM IS (90) DURING MX FOUR BEARS
OF UNDERGRADUATE WORK (100) I PARTICIPATED (25) if (50)
A0TCR (50) DIRECTOR (60) AND (60) TECHNICIAN (70) IN (40)
ALL COLLEGE (45) DRAMATIC PRODUCTIONS (80) SOME OF WHICH
WERE TAKEN ON STATE WIDE TOURS (80) TO (70) VARIOUS HIGH
SCHOOLS (10) AND JUNIOR COLLEGES (100) FOLLOWING THAT (45)
I WAS SatPLOIED (5 ) AS (65) TEACHER OF SPEECH AND ENGLISH
(45) IB A (35) SMALL HIGH SCHOOL
Reading
I THINK THERE ARE ABOUT FIVE POINTS (25) IB MX
BACKGROUND (5) AND TRAINING (95) THAT I MIGHT MENTION (5)
AT THIS TIME (100) ONE OF THEM IS (100) DURING MX FOUR
XEARS OF UNDERGRADUATE WORK (100) I PARTICIPATE AS
ACTOR (35) DIRECTOR (40) A®3 TECHNICIAN (30) IN ALL COLLEGE
DRAMATIC PRODUCTIONS (100) SOME OF WHICH (25) WERE TAKEN
OB STATE WIDE TOURS (30) TO VARIOUS HIGH SCHOOLS (5) AND
JUNIOR COLLEGES (100) FOLLOWING THAT (95) I WAS KMPLOXED
AS TEACHER OF SPEECH A ® ENGLISH (15) IN A SMALL HIGH
SCHOOL
99
Appenaix X
TABLE X (oont*)
Subject E
X HAVE BEEN ASSIGNED A TOPIC (65) Of (1 0 )
3PBAKI1® (20) OB (30) PREPARATION (65) FOR MX FUTURE
JOB (100) HOST (50) FIRST OF ALL (25) WE SHOULD ASK THE
QUESTION (40) WHAT (55) X3 (45) MY FUTURE JOB TO BE (95)
AND (40) AS NEARLY AS I CAN ANSWER THAT QUESTION NOW (100)
X CAN SAT (20) THAT IT WILL (60) BE A JOB (20) HAVING TO
DO (20) WITH EDUCATIONAL RADIO (100) PROBABLY (10) WORK
IN SOME EDUCATIONAL INSTITUTION
8 m u m
1 HAVE BEEN ASSIGNED (30) A TOPIC (15) OF
SPEAKING ON PREPARATION OF MY FUTURE JOB (100) NOW (30)
FIRST OF ALL (15) WE SHOULD ASK THE QUESTION (95) WHAT
IS MX FUTURE JOB TO BE (100) AND AS NEARLY AS I CAN
ANSWER THAT QUESTION NOW (100) I CAN SAY (20) THAT IT
WILL BE A JOB (15) HAVING TO DO (5) WITH EDUCATIONAL
RADIO (90) PROBABLY WORK IN SOME EDUCATION INSTITUTION
100
TABLE X (oont.)
Sub;)wet F
THEREFORE (5) IB PREPARATION FOR IT (100) X
WQUID TRY TO GST (5) EVERS ANGLE (100) OF PRODUCTION (60)
AMD DIRECTION (10) OP FLATS (100) THIS WOULD ENTAIL (95)
COURSES IN SCENIC DESIGN (100) COURSES IN STAGE LIGHT­
ING (100) AND THE OTHER TECHNIQUES (40) OF PRODUCING A
PLAT (100) IT WOULD ENTAIL NECESSARILY (95) ADVANCED
COURSES (85) IN DIRECTING (85) AND IN (30) APPROACHING (60)
THE PROBLEM (95) OF ffWTODOCm A P M *
BatflAitt
THEREFORE (JO) IN PREPARATION FOR IT (100) I
WOULD TRY TO GST EVERY ANGLE (5) OF PRODUCTION (15) AND
DIRECTION OF PLAYS (100) THIS WOULD ENTAIL (30) C0UR3ES IN
SCENIC DESIGN (90) COURSES IN STAGE LIGHTING (100) AND
OTHER TECHNIQUES (10) OF PRODUCING A PLAY (100) IT WOULD
ENTAIL NECESSARILY (15) ADVANCED COURSES IN DIRECTION (100)
AND IN APPROACHING THE PROBLEM OF PRODUCING (55) A PLAY
101
APPEMDIX TO PART XIX
Appendix A
TABLE X.
Pitch lM79.1*
Subjects
Median
Pitch
Leva! *
A
196
B
216
G
219
D
E
F
194
hadlan
Pitch
Level **
21.52
Mean
Pitch
Level *
216
Mean
Pit eh
Level **
21.64
22.31
22.48
22.73
22.20
214
22.16
21.45
197
21.55
22.80
229
%L%
lr\
I -L t
* In cycles par sacend.
** Xn tones above zero frequency level*
.86
X
App@i*&iX B
TABLE
m U m H&S&
►Jects
Lgvola and IgtajL Pitch Ranges.
Median *
Range **
A
196
10.48
B
216
7.47
G
219
14.19
D
214
1 0 .8 6
E
194
1 1 .2 8
F
229
8.49
* Cycles per second*
** Tones*
tor?
TABLE X.
mate* Mlad
M IvMu&l
i p r MSubgit&Ma
Miat M. S M
,1eota.
Subject®
S*P •
A
1*81
B
1*59
G
1 .5 6
n
1*37
&
1*49
f
1.54
Mean S#D»
1*56
104
Aj>p#nAi* 0
®A81* I;
ssmtM
Subjeota
A
B
C
D
M
P
32,61
26.52
2 8 .7 0
2 6 .2 2
25*64
27.17
m
24* T O
23.64
23.93
24,35
23.98
25.58
90
24.13
24.65
24,88
23.92
23.37
24.98
85
23.64
24,02
24.48
23.61
23*19
24.43
80
23.26
23.61
24.08
23.35
22.87
24.19
TO
22.93
23.33
2 3 ,8 1
2 3 .1 1
22.55
23.94
70
22.60
23,09
23*38
2 2 .9 0
22.33
2 3 .6 6
65
22.25
22,85
23.33
2 2 ,6 8
22.13
23.41
60
21.94
22.59
23.09
2 2 .5 0
21.91
2 3 .2 0
55
21.73
22.41
22.79
22.35
2 1 .6 8
23.00
50
2 1 .5 2
2 2 .2 6
22.43
2 2 .2 0
21.43
22.80
45
21,29
2 2 ,1 2
22.33
22.04
21.27
22.60
40
21,04
21.95
22.15
2 1 .8 6
21.08
22.41
35
20.81
21*70
2 1 .9 8
21.65
2 0 .8 8
22.23
30
2 0 .5 6
21.44
2 1 .7 8
21.44
2 0 ,6 8
2 2 .0 5
OK
<o
2 0 ,3 0
2 1 .1 7
21.59
2 1 .2 2
20.48
2 1 .8 1
20
20.04
20.89
2 1 .3 8
2 1 .0 1
2 0 .2 2
21.40
15
19.72
20.60
21.15
20.80
19.96
2 1 .2 6
10
19.37
20.25
20.71
20.60
19.74
20.93
5
1 8 .9 6
19.73
20.42
2 0 .2 1
19.36
2 0 .2 7
100 *
1 05
Appondlx K
TABLE I. 5gc£s2& J*n Z a S S A fit g£j|&S
t J K a M t U M M
fi£
Subjects
A
B
0
D
E
F
Median
1C$
•44
.29
.46
.30
*41
.40
Median
200
•8 9
.65
.95
.64
.84
.79
Median
305?
1*45
1.15
1.35
1.0(3
1.25
1.18
Median
40#
2.04
1.65
1.78
1.46
1.65
1.61
Median
50#
2.63
2.16
2.22
1.89
2.08
2.13
Median
60#
3.22
2.72
2.70
2.34
2.66
2.65
Median
79#
3*92
3.72
3.33
2.81
3.23
3.17
Median
80#
4.77
4.40
4.17
3.32
3.78
4.01
Median
90#
5,79
5.41
5.01
4.15
4.47
5.31
Median LOOpS
10.48
7.47
14.19
10.86
11.28
8.49
Appendix V
TABLE 1;
Mean Extent ef PhonalIona, InflectIona and Shlfta.
Subjects
D
E
B
c
2.7?
2*06
2*66
2.33
2.98
2.87
All
InflootIona
2*06
1.97
2.5©
1.91
2*18
2.19
Downward
Inflections
2*12
2.04
2.60
1.84
2.07
2.24
Upward
Inflections
2# 00
1*89
2 *42
2*00
2.32
2.14
A
Kean Extent ef
Phonatlona
F
Kean Extent of
Inflootlone
Mean Extent at
Shlfta Within Phrases
All
Shift®
1*5©
1• 78"
1.94
1.26
1.66
1.79
Downward
Shifts
2*09
1.79
1.86
1.00
1.43
1.72
Upward
Shift®
1*26
1*78
2*02
1 .4 3
2.08
1.82
2.32
2.46
2*18
1*73
1.52,
2 .04
1.65
1.99
2.85
2*75
2*39
2.54
Wean Extent of
Shifts Between Phrases
All
Shifts
3.54
2*45
Downward
Shlfta
2*87
*9©
Upward
Shifts
3.67
3.08
* Tone*
" v,
1 0 7
Appendix ©
$ABUS
li
Measures o£ Rate of fitch.Change and Humber of
Subjects
a
c
9.66
9.19
11.32
All
Inflectlens
15.41
13.96
Downward
Inflections
15.76
Upward
Inflections
D
1
P
8.40
10.41
8.47
16.76
13.17
18.31
14.63
13.78
15.94
12.97
15.86
15.91
15.00
14* 20
17.38
13.44
21.22
13.05
All Extents
5.91
6.52
5.71
5.74
6.23
6.77
Semi-tone
or Greater
4.51
3.85
4.24
3.95
4.76
4.49
less than
Serai-tone
1.34
2.67
1.47
1.79
1.47
2.28
All Extents
5.50
6.09
5.24
5.43
5.16
5.39
Semi-tone
or greater
4*41
3.72
4.15
4.03
4.19
3.97
less than
Semi-tone
1.09
2.37
1.10
1.40
*98
1.91
A
Kean Rate of
Fitch Change
Mean Rate of
Fitoh Change
During Inflections
Kean Humber of
Changes of
Pitch per second
Including Shifts
108
Appendix H
TABLE Is
Mean a m w r m si
Subjects
A
Words per
Minute
156.6
B
159.5
C
190.2
D
149.5
B
153.5
f
166.0
Total Read*
1
Time
21.08
20.69
17.35
22.08
21.50
19.86
Total Bhon*
ated Time
12.69
13.50
10.85
13.42
13.65
13.58
.60
.65
.63
.61
•64
Moan Ratio of
Phonated Time
to Total Time
.68
1 0 9
Appendix I
tmm
i*
m m
, ffauset and
Subjects
A
B
C
13
£
F
Fhonatlons
*39
.35
.35
.35
.47
.52
All
Fau&ws
*26
.It
.SI
.23
.S3
.25
Pauses Within
Fhraaoa
,09
.09
.03
.11
.09
«IS
Pauses Between
Phrases
.to
.61
.6?
.76
.86
.59
All
Infloat Ions
.15
.16
.17
.16
.15
.SO
Downward
Infloot ions
.15
.16
.17
.15
.14
.19
Upward
Infloot ions
.16
.13
.16
,17
.16
.
in Duration of
SI
Документ
Категория
Без категории
Просмотров
0
Размер файла
7 758 Кб
Теги
sdewsdweddes
1/--страниц
Пожаловаться на содержимое документа