‘ \="13LE ‘ V; C? N k—‘éQRBS 1 a ‘ 0|! '9; .1. Cr‘ .t ”‘G‘J‘.1.".511«1.1C‘2211, 1-1.».31. 9:4 ’ r x '- o 1?;I_"“: (“£1 " m” ‘ ‘11 (31:13:) 16313.2). 1.11 11.11.11.114 ‘2 D“ \- V3 E‘lu.£ DIFFEREECfi ., .1 ”Tl-1's 31" z .313 $313-$911 e‘.’ ‘11 A: rs-m 12111.,cixng'ef“: RH" HW 3113-3 S'Filut; 11111-11131: . .t ‘1 an ." wgé 3", k. 3:131'fi3‘31.i‘i¢iaan 1115: :2. mists LIBRARY Michigan State University ABSTRACT COMPARATIVE ASSESSMENT OF VISIBLE DIFFERENCES BETWEEN VOICED AND UNVOICED WORDS by Richard M. Fulton There seems to be general agreement that the use of voice when teaching lipreading is desirable, however, there seems to be some lack of agreement among teachers of speechreading concerning the extent to which voice should be used. The purpose of this investigation was to determine whether any appreciable visible differences existed be- tween the voiced and unvoiced production of words. This was accomplished with a frame—by-frame analysis of a motion picture film containing six speakers saying four words both with voice and without voice. The variables of lip opening, mouth width, Jaw movement, eyebrow move- ,ment, mouth teeth and tongue area, and the visibility of the tongue and teeth were all analyzed. 3 Data relating to lip opening, mouth width, Jaw .movement, and mouth and teeth area were plotted on ten graphs showing distance in either millimeters or fractions of square inches and time was shown by the number of frames of film necessary to complete the word. Standard deviations were calculated for lip opening, mouth width, Richard M. Fulton jaw movement, and mouth area. In addition, the visibility of the tongue and teeth was determined. However calcu- lations for both eyebrow movement and tongue area were unobtainable. From analysis of the data, there appear to be appreciable visible differences in the size and/or amount of lip openings, mouth widths, jaw movements, and mouth and teeth areas between voiced and unvoiced words. Also an appreciable difference was found in the percentage of time that both the tongue and teeth are visible in voiced and unvoiced words. The present study was exploratory in nature. The data collected suggest areas for further study. COMPARATIVE ASSESSMENT OF VISIBLE DIFFERENCES BETWEEN VOICED AND UNVOICED woaps By 2 1. 1.; Richard M} Fulton A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Speech 1964 N g x. Q; (U cm C“ 5) .3; C) 00 TABLE OF CONTENTS Page LIST OF TABLES . . . . . . . . . . . . iii LIST OF FIGURES . . . . . . . . . . . . 1V Chapter I. STATEMENT OF THE PROBLEM . . . . . . 1 Introduction . 1 Statement of the Problem and Purpose of the Study . 1 Importance of the Study 3 Limitations 3 Definition of Terms A Organization of the Thesis 6 II. SURVEY OF THE LITERATURE . 7 Summary . . . . . . . . . . 12 III. SUBJECTS, EQUIPMENT, AND PROCEDURES . . 1A Introduction . . . . . . . . . 14 Subjects . . . . . . . . . . 14 Equipment . . . . . . . . . . 15 Procedures. . . . . . . . . . 15 IV. RESULTS, ANALYSIS AND DISCUSSION . . . 20 Introduction . . . . . . . . 20 Results and Analysis . . . . . . 21 Discussion . . . . . . . . . 39 V. SUMMARY, CONCLUSIONS, AND IMPLICATIONS FOR FURTHER STUDY . . . . . . . #2 Summary . . . . . . . . . . 42 Conclusions . . . . A3 Implications for Further Study . . . 44 APPENDIX . . . . . . . . . . . . . . 48 BIBLIOGRAPHY . . . . . . . . . . . . . 45 ii LIST OF TABLES Table 1. Mean Per cent of Time that Tongue and Teeth were Visible for Men 2. Mean Per cent of Time that Tongue and Teeth were Visible for Women. 3. Mean Standard Deviations Over Sex. iii Page 38 no Figure IO. LIST OF FIGURES Mean lip Opening and mouth width for six subjects saying zip both with and with- out voice . . . . . Mean lip Opening and mouth width for six subjects saying have both with and with— out voice . . . . . . . Mean lip opening and mouth width for six subjects saying fLul both with and with— out voice . . . . . . . . Mean lip Opening and mouth width for six subjects saying naught both with and with- out voice Mean jaw movement for six subjects saying zip and have both with and without voice Mean jaw movement for six subjects saying fLul and naught both with and without voice . . . . . . . . . Mean mouth area for six subjects saying zip and have both with and without voice . . . . . . . . . Mean mouth area for six subjects saying fLul and naught both with and without voice . . . . . . . Mean teeth area for male subjects saying four words voiced and unvoiced Mean teeth area for female subjects saying four words voiced and unvoiced iv Page 22 24 25 26 28 29 CHAPTER I STATEMENT OF THE PROBLEM Introduction A major problem, in the teaching Of lipreading, involves the proper utilization Of auditory cues. If the subject is deaf, the use of voice is not a matter for con~ sideration in the acquisition of lipreading. If the individual is hard Of hearing, however, the use Of voice may give the student auditory cues which he will rely upon instead of visible cues or a combination of visible and auditory cues. There seems to be general agreement that the use of voice is desirable. However, there seems to be some lack of agreement among teachers of speech- reading concerning the extent to which voice should be used when teaching lipreading. This study will attempt to indicate the differences that exist between the voiced and unvoiced production of four words: zip, have, foul, naught. Statement Of the Problem and Purpose of the Study The purpose Of this investigation was to determine whether any appreciable fisible differences existed between the voiced and unvoiced production of words. I This was accomplished in part by a systematic evalu- ation of the movements of various facial structures through a frame-by-frame analysis of a film containing six speakers each saying fourteen words with voice and the same fourteen words without voice. In all, nine measures and five judgments were made on each frame. In the frame-by-frame analysis the measure- ments made were to determine the amount of movement of the jaw, lips, mouth, and eyebrows. The judgments were to indicate if the teeth and tongue were visible. The entire face from the chin to the tOp of the head was included in the filming. The following questions were asked with regards to whether a difference existed between the voiced and un- voiced production Of the words. Is there a difference between: (1) the size of lip Opening at the philtrum? (2) the size of mouth width as measured from one corner of the mouth to the other? (3) the amount of jaw movement as measured from the apex of the nose to the center of the chin? (A) the amount Of eyebrow movement as measured from the apex of the nose to each eyebrow and then between eyebrows? (5) the amount Of Open mouth area, teeth area, and tongue area visible as measured by a planimeter? (6) the visibility of the teeth and tongue in male and female subjects? Importance of the Study It was thought that this analysis of facial move- ments would assist in resolving a long standing question as to the importance and/or advisability of using vocal or subvocal speech while teaching speechreading. Limitations The small sample used in this study was itself a limitation. The study was delimited to four words and six speakers. Originally each of the twenty-eight words said by a Speaker was to be subjected to a fourteen point evaluation. This, however, proved impractical due to the time limit at hand. I The speakers were three men and three women. In addition, the sounds used in the selection of the four words, and the proportion in which they appear in those words, was to be limited to those observed by Dewey as reported by Miller.1 However, in the final analysis it was not possible to have all of the sounds used in the English language represented in the four words chosen as a sample. 1George A. Miller, Language anq_Communication (New York: McGraw-Hill Book Company, Inc., 195I7, p. 86. Definition of Terms Appreciable movement.-—That movement or physical change great enough to be measured or estimated. Word list.-—Fourteen randomly selected monosyllabic and phonetically loaded words.2 (See Appendix A, on page 49.) Subjects.--Three female and three male college students, trained in Speech and hearing science, filmed while Speaking the fourteen words. Sixteen millimeter plus-X reversal film.-—B1ack and white silent single system motion picture film. Auricon sixteen millimeter camera.--Motion picture camera with a 102mm lens employed for filming. Frame.--Section of film 5/16 of an inch. Twenty- four frames equal one second of filming time. Head guide.--A clamp fastened to the back of a chair to prevent movement of the heads of the Speakers. The construction was such that it did not inhibit free jaw movement. Rear projection screen and mirror.--The screen on which the film was projected by way of a mirror set at a 45 degree angle. The screen was constructed Of a double thickness of plate glass with a piece of frosted H I! 2All the words were monosyllables except ago which has two syllables. acetate (frosted side out) stretched over the glass. The clearer image appeared on the frosted side of the screen. Bell and Howell sixteen millimeter movie projector.-- Projector used for projecting filmed images onto the rear projection screen by way of the tilted mirror. The film was advanced frame-by—frame. Dp£§,—-A black dot was marked above the eyebrows of each subject, on the apex of his nose, and on the center of his chin. These were used as reference points when measuring facial movements. Vernier clappers.—-Clappers used to make linear measurements from one fixed point to another fixed point. Measurements are reported in millimeters. Planimeter.-—A sensitive measuring instrument used to measure the area of the teeth, tongue, and mouth made visible while saying the word list. Units of the measurements are reported in square inches. Anatomical terms Philtrum.--The shallow groove in the center of the outer surface of the upper lip. Apex.--The tip Of the nose. Crown.-—The portion Of the teeth projecting above the gum. Organization of the Thesis Chapter I, which is introductory, discusses the purpose of the study, states the problem, lists the questions that are raised, considers the importance of the study, the limitations, defines the terms, and outlines the organization. Chapter II consists of a survey of the literature pertinent to a study of the voiced and unvoiced production of words. Chapter III describes the subjects, equipment, filming procedure, and measuring techniques employed in this study. Chapter IV discusses the analysis and results of the study. Chapter V contains the summary, lists the conclusions, and the implications for further research. CHAPTER II SURVEY OF THE LITERATURE This chapter considers literature pertinent to the present study of voiced and unvoiced production of words while teaching speechreading. Teachers of the deaf and hard of hearing appear to favor the use of voice as the best approach. The feelings expressed in the literature can be categorized into two areas: those strongly favoring the use of voice and those strongly against using exaggerated and slow Speech. A majority Of the teachers and educators found in the literature lean toward the use of voice while teaching lipreading, (l, 2, 3, 4, 5). l Bruhn feels the teacher can assist the student in learning to speechread by: . . . making the work in the class room as nearly as possible typical Of the speech a person encoun— ters outside. It is therefore obvious that speaking without voice should never be permitted. If a person with a fair amount of hearing is taught to read lips when no voice is used, he will demand the same conditions outside. The moment he hears the voice he will strain every nerve to hear, rather than watch the lips. The habit is too strong. 1Martha E. Bruhn, The Mueller—Walle Method of Lip Reading for the Hard Of Hearing (Boston, Mass.: M. H. Leavis Pub., 1949)] p. 16. 2 Berry and Eisenson concur with Bruhn but are more specific when they say, "The teacher may begin Speech— reading with words and phrases employing mainly visible articulatory movements and normal rate Of speaking. Varying degrees of loudness Of voice are used by the teacher, but at no time is lip movement exaggerated, even when the tone is inaudible." the Newby3 agrees and states: Some voice should be employed during the Speech— reading lessons, so that the patient can make use hearing, as well as vision. It is preferable, therefore, for the therapist to speak in a voice which is barely audible to the patient and which presents a difficult listening situation for him. Macnutt4 has some additional thoughts on how audible instructor's voice should be. Voice should be used throughout the lesson and, so far as possible, of the same intensity that one sould use in Speaking to a normally nearing person at very close range. The hearing acuity of the pupil who hears most will determine the amount of voice that can be used. The pupil who hears most will determine the amount of voice that can be used. The pupil may hear the sound of the voice, but should not hear enough to understand all that is said without watching the Speaker. Neither the teacher nor the pupils should whiSper or speak with- out voice in the lipreading class. 2Mildred F. Berry and Jon Eisenson, Speech Disorders (New York: Appleton-Crofts, Inc., 1956), p. 4641 3Hayes A. Newby, Audiology (New York: Apple- ton-Century-Crofts, Inc., 1958), p. 277. uEna G. Macnutt, Hearing With Our Eyes (Boston, Massachusetts: Author, 1952), p. 2. Faircloth5 feels that the only time voice is not used with speechreaders is during a test. There is one way of putting a group of lip- readers on a level for test, and that is by Speaking without any sound Of the voice. It is the only time when lipreaders Should be Spoken to Silently. Ordinarily when addressing a lipreader, a Speaker Should use a natural tone and enunciate clearly. A number of authorities (6, 7, 8, 9, 10, ll, 12) indicate that the use of either exaggerated or slow speech when teaching speechreading is not recommended. Stowell6 points out that teaching lipreading without voice should not be done. . Speak naturally in tempo, in phrasing, and in movements of the lips. It is a great mistake for teachers to try to aid their pupils by exaggerated lip movements and by unnatural phrases. Teachers of lip reading find that if they give their exercises without voice, there is a tendency toward exaggeration of lip movements. The Kinzies7 agree by saying, ”Slow utterance in instruction or practice for the purpose of aiding the student is a mistake which, if persisted in, works incalculable harm." 5N. Faircloth, Lip-Reading Study and Practice (Toronto: The Ryerson Press, 1946), p. 51. 6Agnes Stowell; Estelle E. Samuelson, and Ann Lehman, Lip Reading for the Deaf Child (New York: The Macmillan Co., 1928), p. 16. 7Cora Elsie Kinzie and Rose Klnzie, Lip-Reading for the Deafened Adult (Chicago: The John C. Winston Co., l93l), p. 28. IO The above authors are supported by Montague8 when she says that sentences, ". . . may be Spoken smoothly and rapidly. In fact, it is very important that they be said naturally, aS in conversational Speech, and not pronounced slowly, word by word." Bruhn9 also feels that the teacher Should be careful in her manner of Speaking by stating: It is important that the Speaker should Speak distinctly, yet avoid all exaggerations. Speaking in a loud voice, or with the head lowered, or emphasizing each word with a gesture of the head must be avoided. Pedreylo further agrees that ”mouthing" is to be avoided: Do not ”mouth” or exaggerate the movements of the lips. Some teachers do this in an effort to help the student. This only tends to make normal Speech con- fusing and difficult to understand. Use a moderate rate Of speaking. Nitchiell gives a rationale for not exaggerating when in her discussion of lipreading she says: In the second place the method should aim always to study or see the movements as the words are pronounced Quickly. It is true that it would be easier to See them when Spoken slowly, but it is also true that to produce the best results the eye Should be trained from the first to see things as they must always be seen in ordinary Speech, and that is rapidly. 8Harriet Montague, LipiReading Lessons for Adult Beginners (Washington, D.C.: The Volta Bureau, 1945), p. 5 9Martha E. Bruhn, The Mueller—Walle Method of Lip— Reading for the Deaf (Lynn, Mass.: Thomas P. Nichols and Son Co., 1920), p.5. loCharles Pedrey; LeRoy D. Hedgecock, and Waring J. Fitch, Manual of Hearing Therapy,gThe Teaching of Lip Reading N. P. I(l947), p. 6. 11Elizabeth Helm Nitchie, New Lessons in Lip Reading (New York: J. B. Lippincott Co., 1950), p. 27. 11 Finally, Newbyl2 summarizes the above thoughts by saying, "Talk naturally. Don't exaggerate your articula- tion, and don’t speak carelessly or too rapidly." The American Hearing Societyl3 tries to define the type of "voice” that one should use in the following material. The Inaudible Voice: To be used whenever we want the child to depend upon his lip—reading to a greater extent than upon his hearing. This voice is the one most used, most important, and most difficult . . . It must not be a whiSper; first, because a whiSper can carry almost as well as an audible voice, and secondly, because we have discovered that the tendency to mouth and exaggerate is great when we adOpt such a technique as whispering. Nor is it a voiceless articulation of movements, because mouthing and exaggeration are almost invariable accompaniments of this strange technique. Rather is the inaudible voice an ordinary, but very low conversational voice. A third aspect, of the problem of whether or not to use voice in teaching lipreading, involves the use of voice but eliminates both exaggeration and auditory cues. O'Neill and Oyerlu feel instruction should be divided into two stages. . . the initial stages of aural rehabilitation give training without voice SO that the hard- Of hearing person can focus his attention upon the visual aspects of Speech. 12Newby, loc. cit., p. 246. l3American Hearing Society, New Aids and Materials for Teaching Lip Reading (Washington, D.C.: American Hearing Society, 1943), pp. 18-20. 1“John J. O'Neill and Herbert J. Oyer, Visual Commu- nication for the Hard of Hearing (Englewood Cliffs, N.J.: Prentice-Hall, Inc., 1961), pp. 74—75. 12 Once he is aware Of the visual channel, training can then move toward the utilization of the combined auditory-visual modalities. One method Of accomplishing this is to divide the therapist from the Observer. Pedrey, et a1.15 feel that: Ideally, lip reading is taught in a room which is divided by a glass partition. The class is thus separated from the therapist, and the therapist is able to Speak naturally without having his voice heard by the class. However, there are relatively few such situations, and the therapist with his class are usually in the same room. Davis and Silvermanl6 support this method by stating that practice Of speechreading: . . . with no sound is given by the use of a large glass partition between speaker and Observer. The speaker converses behind the glass in normal tones, with no distortion resulting from attempted muffling, and the Speechreader discovers that he has followed the conversation, deSpite the complete absence of auditory cues. Summary This review Of the literature dealing with the voicing and unvoicing Of words, while teaching lipreading, brought several facts into focus. First, a majority of the authorities in the area Of the deaf and hard of hearing prefer to use voicing. 15Pedrey, loc. cit. l6Hallowell Davis and S. Richard Silverman, Hearing and Deafness (New York: Holt ,' Rinehart and Winston, Inc., 195253 p. 362. (Chapter 12, Speech Reading", M. D. Pauls, Ph.D. . 13 Secondly, the literature made clear that any techniques which cause the teacher to either exaggerate or Speak slowly during a lesson are unacceptable. Lastly, a suggestion was Offered by several authors on how to allow the use of voicing and to control auditory cues at the same time. CHAPTER III SUBJECTS, EQUIPMENT, AND PROCEDURES Introduction The problem in this study was to determine whether any appreciable difference exists in the facial movements of Speakers as they Speak words with voice and then with- out voice. A film was made with Six subjects saying four— teen words with voice and the same words unvoiced. Move- ments involved in saying these words were then analyzed frame-by-frame. Subjects Selection of Speaker subjectS.--Six individuals were selected as speaker subjects. The basic criteria employed in the selection Of the six subjects were sex and training as a Speech therapist. None Of them had any knowledge of what was being attempted in this study. The three men had general American dialects while two of the three women were from the South and one from the East. All of the subjects were graduate students in the field of speech and hearing. 14 15 Equipment The following equipment was employed in this study: Motion picture camera (Auricon l6mm., Ser. NO. 32328, Lens 102mm.). Film (Plus-X reversal, Silent, black and white). Lamps (conventional standard flood lamps). Chair (with head clamp attached to the back). Cue cards (fourteen 6 x 3 1/4 inch cards with one monosyllable word each). Word list (six lists of the same 14 words each arranged differently). Movie projector (Bell and Howell l6mm., MO. 173, Ser. NO., 41829). Rear projection screen and mirror. Vernier clappers. Planimeter (Ott Planimeter, Ser. NO., 94064). Rule card (a 6 x 31/4 inch card with two inches marked Off on its lower edge). Procedures Selection of monosyllabic word list.--The word list1 employed in this study consisted of a random sample of English monosyllabic words. The choice of words, however, was based on phonetic content, in keeping with the findings Of Dewey as reported by Miller.2 1See Appendix. 2Miller, loc. cit. 16 The word list was read twice, once with voice and then again without voice. Each Speaker said the word list in a different order,3 to provide a variety Of sounds before the various words. This precaution prevented a stereotyped production of the word lists which could conceivably bias the measurements. Four words (Zip, Have, Foul, and Naught) were chosen from the word list to be evaluated. Thus each word was spoken twice making a total of eight words from each subject, or a grand total of forty—eight words to be measured and compared. Filming situation.—-The filming of the Speaker sub- jects was carried out in the Michigan State University Audio-Visual Film Production Studio. The subjects entered the studio one at a time and were seated in a chair with a head clamp attached to the back. Each subject's head was placed in the clamp which, however, was not tightened. Their heads were positioned between the sides of the head guide simply as a reminder for them not to lean forward. This action, however, seemed to restrict the amount of eyebrow movement, facial expresn sion and head movement generally evident while speaking. Before filming each subject, a black ink dot was placed on the apex Of the nose, center of the chin and 3See Appendix. 17 above each eyebrow. The dots remained on the subjects during the filming. The dots were used as reference points when measuring facial movements. The distance from the subjects to the camera was kept constant. (See Appen— dix, page 50, for an illustration Of where the dots were located on the face of subjects.) Lighting was supplied by conventional standard flood lamps. The intent was to have the subject's face appear as natural as possible and for the film to record only the surfaces that would ordinarly reflect light. This same light would be all that a lipreader would have to assist him. After the subject was positioned, and the camera focused, a few frames were filmed with the rule card held below the subject's chin. The investigator stood to the left of the camera and held up a cue card on which a monosyllabic word was printed. Each cue card ’ had a number on its back which correlated with the posi- tion Of that word in the word list order selected for that particular subject. After the subject said the first word, the second wOrd was held up and so on. This same procedure was repeated with each subject as he or she said the fourteen words with voice and then without voice. The directions given to each Speaker were as follows: Sit down and be comfortable. Please rest your head on the post behind you. Relax. I will Show you fourteen words which are to be read by you and 18 Spoken out loud (or Silently). You will start from a state Of rest. Say the word and then pause, with your lips closed, and say the next word. Each subject was thanked and asked not in inform the waiting speaker subjects of what had just taken place. Measuring situation.--The measurements were made in the Visual Communication section of the Speech and Hearing Science Laboratory in the Department of Speech at Michigan State University. The film was projected onto a rear pro- jection screen by way of a mirror tilted at a 45 degree angle. The rear projection screen was placed parallel to the floor and rested on a support, so a planimeter could be employed on a flat surface. Life-Size images were projected. This was accomplished by measuring the distance between the two, one inch lines marked on the rule card, which had been placed under the subject's chin at the beginning of the filming of each list of words. The first measurement was the distance from the inferior surface of the upper lip at the philtrum to the superior surface of the lower lip. Next the distance from one corner Of the mouth to the other was measured. Thirdly, the distance between the dots above the eyebrows and the distance from the apex of the nose to the dots above each eyebrow was measured. The Sixth measure was from the apex Of the nose to the center Of the chin. The above measurements were made by using a vernier clappers. 19 The next three measurements were made with the aid of a planimeter. This included the area of the Open mouth, tongue, and teeth. When the teeth were visible, but apart, a separate measure was taken for both the upper and lower teeth. The pole arm of the planimeter was set at 44.02 and the tracer arm was set at 5 — 8.05. Five judgements were made by looking at the picture and observing whether or not the tongue was visible and if SO whether it was positioned up, down, or out of the mouth and whether or not the teeth were visible. Often this writer could detect the presence of the tongue and/or teeth but would be unable to measure them with the plani- meter because of a lack of form on the part of the structure involved. Measurements were made as the subject's mouth started to Open and finished when the mouth had closed and the lips were together. All of the measurements and judgements were recorded on an especially prepared form. (See Appendix on page 51). CHAPTER IV RESULTS, ANALYSIS AND DISCUSSION Introduction Physical measurements were made in an effort to answer the six questions set forth in the first chapter. This was accomplished by a frame-byrframe analysis of a motion picture film. The following questions were asked with reference to whether a difference existed between the voiced and unvoiced production of words. IS there a difference between: (1) the size of lip opening at the philtrum? (2) the Size Of mouth width as measured from one corner of the mouth to the other? (3) the amount of jaw movement as measured from the apex of the nose to the center of the chin? (4) the amount Of eyebrow movement as measured from the apex of the nose to each eyebrow and then be- ween eyebrows? (5) the amount of open mouth area, teeth area, and tongue area visible as measured by a planimeter? (6) the visibility of the teeth and tongue in male and female subjects? 20 21 Results and Analysis Lip Opening and Mouth width.--Figures 1 through 4 present the average amount Of the lip Opening and mouth width of six subjects, divided into subgroups of men and women, for both voiced and unvoiced words. The intervals along the ordinates represent distance measured in millimeters. The abscissas represent time measured in frames with twenty-four frames equaling one second of filming time. It was necessary, because of a difference in the duration in the pronunciation of the words by each speaker, to indicate, where appropriate, what segments of a graph line represent a mean of three or two Speakers and where the data used is based on only one Speaker. The symbol (12) was used tO indicate a mean of two Speakers and (11) for that of only one subject. §_p,--In Figure l, the male subject's unvoiced line Q___ .____) exhibits greater flux in both the lip Opening and mouth width measurements then the voice line (____). They both, however, follow the same basic contour with the unvoiced line generally above the voiced line. This process is reversed in the case of the female subjects. Their voiced line (. . . .) is usually above the unvoiced line (- - - —) which appears to follow the contour of the men's unvoiced line more so than the women's voiced. Thus, the female's unvoiced line also diSplays considerable .OOHO> psonpaz cam Spa: neon mHN mafihmm mPOOnQSm xam mom Suva: cpsoa can wcflcogo QHH Cmoznu.a omswfim NMNuuuuMuuuuuguamuuqfiquhwM§fiiifi~ \N’o \ . 28mph. .n. z a l/ I ‘ 00/ \’ h l h \ I 1‘5 ~ i O o 50 \\ / a & onu‘gog J ..J\\.\/A/ 34 s 11 .. a 0.. oJI-komln H ./% u\|| o Logos ‘1 000““. N- oooooo .fl— \ o\ l I V Q§l . / IMO. \ I «was 3331.11 m L n]. on [It ' '1! Q 7 16-! ccccc O Q a . 0d 0 9 d O a l a. we .9“ 3:3 fi 22 . «5‘ is. 3 E. a; .’ K s v :4: W?!” w L$IE$J§1$J$13 lg] “EL \\. 6“; ‘.‘.‘\\ .. . I . \ o .\ #4 I Q \ 0". t“ I . u/ \ . vb.) ‘0 I” .0 o ’ 4. g, . . as. 1% 23 variation. Finally, there is a very close mirroring of the lines within each category and between the cateogires up to the fifteenth frame where the curves no longer represent a mean of three, or in some cases two, speakers and considerable divergence takes place. H§y§,--In Figure 2, the relationship between curves is essentially the same as those in Figure l. The major difference arises in the mirroring of lines within each category. The curves of the female subjects Show greater Similarity than those of the male. Epgl.-—In Figure 3, the voice curve of the male subjects tends to be superior to the unvoiced curve. Their curves are similar between categories but demonstrate a wide divergence between the seventh and twelfth frames within each category. The women, on the other hand, present a reversal in the position of the lines between groupings. In addition, the women's curves are more alike both within and between categories. Naught.-—In Figure 4, the male subjects' curves follow each other very closely in the mouth width group- ing. However, there is greater separation between curves on the lip opening group. Once again the female subjects provide a reversal in the position of the curves between categories. Lastly, there is a distinct lack of relation- ship between the categories. Lip Opening presents a greater fluctuation then the mouth width grouping which appears level and stable. 24 .OOHO> usozpflz Ono zpflz cpop o>mfl wcfimmm mpothSm wan som Sana: npsoE one wCHQOQO QHH Cmoznn.m osswam M 3'1) «N Pu/QN SEE »\ mu Q Q Q = ex 9 Nu. w / t — mmeaHm ”1 I/oflo . ~ 1 . JMJmAPAM\)/. \\ h. . o . 1 .. /x/. a. a $.5an L...» at: d s. o / - .\ \ Ala- P. . cs . /|I.\N..7 I fio\. 1.45 \.. W L o o a I . .0, \ . 7 M? \UU\O oOoOCOOOo‘oooo\ / N “Cd. \ a J . § {3}. c n I / o- o . \ q. 3.. / Co I. U. c\ \ o L o 95 C, U. o\ \ o :L /'o ‘ 5‘0 /’\\O on \ O O l m >¢ I” s \U/Usl” noon-c unno\o\9-o tn... .95. Mai: 1.11 - m. 1 x mm /... \ 2. /. .. . N1 W/ILIII. cl 2 / a. . VB“ n/ .2, 3o<< \\\. .Jnm I. / s o {IFV. {kt . 0-0” H .. /. 4: .. om» . // \l/ E.\ \\W i . x // .13.... x. it? s o I], \\ / Ec\ 1011 \o . A ‘ . IIIII ..\. 11o: \.. a 3 o. . N / o 4 in‘\ - Cu :1 as 0 o .0 o a. Q 01(- ~ a OK No“ \ on A ../.\r\..\........../ u :5 \ . 91 *3...) 10%.. I m. «2.) \ on as; oo/‘thFa’loooooo\A\\\l~co n s I'.’ nl \0 I. U Q usnflx. 25 .moHo> pzonpaz cam spa: Swen H50% wcahmm mpoanSm me How Suva: Space cam wcficmgo QHH :mmzuu.m mhswam mmEmhm t. ’0 lo to%lowoh a w M h w M: ‘— m 4”! ‘u h o o I d .l o O ’ g... ./0 O l . \\~.Il ll/ ‘6 8’ §/. s.\ \05H .v... Ill 33. . “f 99 so I In.” - g \ I, L «F‘smko m...“ ..~.....w.~...: :lll/ /. \. .. \\ mi u..na\.s.o.i..on». . ~ . ll /\. .. \bflfi 9 D“ 9. 5 \ m .. I . x U adulolo Po... \H” K} .H D. O m ‘.Un.u|° IOr-sj 4 a. . D. IN‘ 0.. L IN a). olll‘u‘IlI'blallo \ s A . . . ._ 1...... it / . . \MF. s“... . .....s u . \. \mfifiw oQQIo‘roowfil‘ "‘h “.1010," ‘OU.~.‘ 03g» . c ‘. ~,‘3 z?! o o \ m “I K / \\ ~ V0 .0.§“/ll \\/.I s ”—9 N I / $5 \ .l- o a r. o o \\ Riki v. . \ p . x55. «5...? :11. x x it?“ 26 .moHo> usonpaz cam spa: cpon pQMSmc wcammm mpothSm xam pom Sana: SusoE new wcflcmgo QHH Cmmznu.: mhstm mmEmhmk\h\uxb>\h~d§o\mNbuhwhws / \/ box .39? KG o .\ uto'o.ouooou.\fl .\ .- . .. 1.533 ull ..\ o3.§> \“~ “fill §.\oo.\ \ II 4N9 . / 02¢§§3o .. \ \ /v\ . .lfi.r\ . . . $.23. «szoi T .1”? - -.- 37IVU '12 l+l I J 27 Jaw movement.--Figures 5 and 6 present the average amount of jaw movement of six subjects as previously set forth except the Jaw movement of two words are given simultaneously on the same graph. The ordinates and obscissas remain the same. Zip and Have.-—In Figure 5, the male subjects on the word zip have parallel curves with the unvoiced curve proving predominate. In the word have, however, the curves do not mirror and here again the unvoiced curve is higher. As a contrast, the women's curves tend to mirror one another with the voiced curve in prominence for both categories. Furthermore, there is a lack of similarity between the curves of male and female subjects in both categories. The female subjects' curves tend to be more peaked while the men are over all higher. Foul and Naught.-—In Figure 6, curves of both male and female subjects show reversals of predominance of voiced and unvoiced curves over each other. Both groups of subjects present lines which are smooth. The curves mirror one another more closely in the word fgul. As in Figure 5, the male subjects' curves are higher then those of the female subjects. Eyebrows.--The amount of eyebrow movement both laterally and vertically exhibited by each speaker was so insignificant that calculations were not practical. Such a lack of mobility was apparently caused by the head 28 .ooHo> psonuaz cam Spa; npop o>m£ paw mHN wqahwm mpoonQSm me pom pCmEo>oE zmw Cmozuu.m ouswam -Wuuuuomfl_§q.:u~flfl.u~fi=8»wnlfimsyu~- .../ moEmpm ... rd H— \o\-\\ :5. \ '0 Nolv~h§l| what? \\. ..\. m 1 up \2 owes. a... or“... a. Vooosooov‘UW: *“0’ H .\x w WK In '.’/ 6 I O I, .0 - I, .l ,1 .\ I, IN 0. >fifl”.’"-h 06 ’3’", to. / 30‘ A 3...»... ti... . / . 0000.. 00 ’ 0.. .. m > < m . /o /O (I.) i'l‘.l\ / / .\ ./\ . /I . reset: sci.\ 29 .ooao> pSOSuHS ocm Spa: :pon ucwsmq cam H509 madman mpoomnzm me how pdoEo>oE wa Cmoznu.m madmam 00‘. O - o. o- co \ll/ . on OK. amassmqmt 2 wwjx £22 zaw a n u m s w. WNJ O 0 lo}: \000 . z /I . . o I xx ooo I‘lfiufi“ to .H D 0 pm . \0 Nb. //\ O 9. o 0 III“, C. \\ FCC—\SNU . o ~ o‘c‘WCoQVoooooso III] o... \ WNW — . 3:} o c a. oo o .Jol.J so . \ \ av / d 7.5»... .. i... \ .0 am .5. c% \I". \ . ygog \I\ an“ / . «g Q ' 0/. /\. er M. r . a... 3<o< 3ND .y. r: Rocco-9‘0 d-OQOO .ooaooo k so Ra o o O\O\§//II ‘00“. u o. \\ 4 3O guide. Each subject positioned his head between the sides of the guide and rested his head on the back of the guide. This was done in order to keep a subject's head from moving forward or laterally; only in this manner could the distance between the camera and subject be kept constant. The result was an immobile head and a lack of expression which reduced eyebrow movement to either a bare minimum or nothing. Mouth area.——Figures 7 and 8 plot the average mouth area of six subjects, as previously set forth except the mouth area of two words are given simultaneously on the same graph. The intervals along the ordinates represent distance measured in fractions of square inches. The abscissas represent time measured in frames with twenty- four frames equalling one second of filming time. Here again symbols are used to indicate when the mean number of speakers represented by a graph line vary from three. The female subjects tended to show greater variation in initiating and concluding words. Therefore, a larger percentage of their graph lines are based on a mean of two subjects than the men. An Ott Planimeter was used to measure the mouth area. This instrument uses a Vernier Unit of measurement which must be converted into square inches. The plani- meter was calibrated and the pole and tracer arms set so one quarter of a revolution equaled one square inch plus 31 or minus two percent. This meant that one square inch equaled 250 divisions on the planimeter's scale. There- fore the following formula can be derived: 1 sq.in./division x division = sq.in./div. l sq.in./250 x 1000 = .004 sq.in./div. .004 sq.in./div. x24 VU (Vernier Units) 0.0§5 sq.in. This formula was applied when calculating the distance in square inches for both mouth and teeth areas. Thus, a distance in fractions of a square inch was obtained. Zip_and Have.-—In Figure 7, the male subjects un— voiced curves tend to be higher then the voiced curves in both words. There is fairly close mirroring in zi , but the men's unvoiced curve in hale exhibits considerably more flux then the women's unvoiced line. The female subjects' voiced graph line is higher then their unvoiced line. Once again, as in Figure 1 through 4, the female subjects' graph lines are predominate to those of the male subjects. In addition, a close comparison can be drawn between the lip Opening and mouth area graphs of both words. They appear to have the same basic contour. Foul and Naught.——In Figure 8, graph lines for fgul_ present an inter—mixing and reversing of position not unlike those found in the lip Opening graph of the same word. However, in naught the curves resemble those lines found in the mouth width graph of the same word, the major 32 .ooHo> psonuaz cam Qua; Spon m>m£ cam mam wcfimwm muomnnsm me pom mohm cpsoa Cmmznu.w mhswam n“ mmewchH .Voa .‘.|I|u.||.gl\ I. . .\\QJy}S§a.S}w nQM... ~ m..\.\ w § 0 o I '0 C. . III I.\ \nn. I I‘luwg lR.V¢o\§\.d. o o In y x nH H N 0. //\N a“. covoo‘mroooo ’ o. 06. o wnms08 ‘0 y. d o. .o W as. ‘0 a a. 'a 4 W. a Elk... a (at 5:0: .. {LII/.1]. ( ( o\¥‘ L 0- ////od 1 \ Pu // a n. . .Voo I o W—I. .l I softd‘ ”O. " I I .y .~ //\- J / / dovu‘\dm.. \oNz. is «.8; Isle“... . on // I I . 4. to} x... . . . . .ss. ... . I \ o I do. a/' I d \ R\ Q N .~¢ 1’ ..|\ \ so. "I - /obo \ core-”luv 2“,\ o n m > < m 0/\ '0’ ’ \ o. o .o fl. n. \s. or ’00. o (no 33 .ooHo> psonvaz cam zuaz npon pnmsm: cum Hoom wsahmm muomwQSm me sow mopm EpsoE Cmozun.m opswfim mesmsmowtufljjfltfldo \czumrqusn«~o . an . . . a... o oo‘voovooo .Iu EVN“. o... c all-NJK .00 \\ ".4 o \6‘ \“fl! d‘ I ll... .000...o|o1u|""’ 000 ‘\ y it. \\ ‘.w\:‘:' 'n' '\|\‘ no.00D90r'00, \ ’. J a «L B m U D < Z W: . ”as. a. L ‘2 L 34 difference being a greater amplitude and a closer paralleling of the lines in the mouth width graph. Teeth area.—-Figures 9 and 10 present the average tooth area of six subjects as set forth under mouth area. The only exceptions to be found between the mouth and tooth area graphs lies in the division of the graph into three areas, instead of two, and a discontinuing of the use of symbols to indicate the mean number of subjects forming a graph line. There are two graphs, one each for men and women subjects. Each graph in turn, is subdivided into three sections: lower teeth, upper and lower teeth, and upper teeth. All four words, in both voiced and unvoiced form, are represented, by a different set of lines. The "both teeth" section has the greatest amplitude with lower teeth and upper teeth following in order. ‘Mgn.--In Figure 9, the upper teeth diSplay a very close relationship between all words. In contrast, the "both teeth” section presents graph lines that are wide apart and scattered. The voiced curve of zip L___) is the only line that goes across the graph. This same feature is found in the lower teeth who's lines are predominatly mixed and interrelated. flgmgn,—-In Figure 10, the upper teeth graph lines show only a close relationship and are more prominent then in Figure 9. There is a greater variability in the 35 .pooao>cs UQm pmoHo> mppoz Loom wCHzmm muomeSm mHmE sow mopm Snoop cmmzuu.m mhswam \l‘ mmEmhm 9.. fix NA/«wx WW@ x: M\ w Qx ox % . N x J o 0} 334!!! {I0 5 ll 14% 5'3wa 234mg: J! Iflgljfiflf/ Es zmz mom Ema momma ITI a» 4.. illlva'lxlla to...“ o o.“ """ ' l/ I ' ‘ I -‘Q' 0" O. .. wk «UK «it: a... 36 .cooao>:: cam cmoao> mcaoz hzom wCHhmm wpoanSm onEmm pow mopm cpmop amozun.OH omswam flew Eon .0 O meefimxulolmttumwxwws/Qfi:uxmhhuhwmw. S5 oSo pooao> mosoz Loom wSHhom mpoowSSm onEoS pom oomo Spoon Sooznn.OH omswam meeehmxunad‘ 2 2J2 @ E Q j... S m a m u m, u m N . Iz/VII/ >1: ..... ii. A ... < I ./\. \§ , ..~......\.u.\..\/./. \ .0 .. filmyfit.1luuVLWfiK4:%. J : Ilo' \. o ‘ 9 /4 \..I 7/. 3 IIIII . £\./ .M :00... / \|\ a ll¢|lldt \\ . . T§I-I $N «hawk. (030% /\. t I Tu...) III. III III III I I Show Em " a .. Lv 75 n V v a ZMEO3 EOE em ewe. sow Koo Koo Rom mmm moo moo Ros Rom 3N .oo>Sb .oo> .oo>SD .oo> .oo>SD .oo> .oo>SD .oo>. .oo>SD .oo> SpooB p50 czom can ostoB zmzoz mom mqumH> mmmz momma oza moozoa some mafia mo BzMommm zamz m mamas ems.ms amm.:m Rmm.o ems.m Rom Rom mm.ma Rm.oa Rmm Rm: zoos ozohu om om oo oo mm Hm ma mo so mm onwsmz mm mm mo oo mm mm mm am so we doom om so mm mm ms mo oo so an op osom Row also woo woo emm moo moo moo RR RNH 3N . OO>CD . 00> .OO>CD . 00> . OO>CD .OO.> .OO>CD .OO> .OO>CD .OO> Spooe uso Szom an oanoE 2m: mom mAMHWH> mmm3 mammfi 92¢ mDoZOB.Hmm ham UoOHO>SD UoOHo> UoOHO>SD ooowo> UoOHo>SD UoOHo> UoOHO>SD UoOHo> moh< Spsoz pSoEo>oz sow coca: nose: wSHSoQO QHQ xmm mm>o mZOHBMQ deQZSBm Z BOZUDM mmc—«zmmz MNEIJIZQMOB I—‘HfllicE-i MMMFIJDSOZ MHUEB mwmmmoz Down Teeth Area Both Q. :3 BOZUDLTJ LTJSL=JZE4 EODE-HI‘. 231-45393: HHS. 0941:1512sz mm