Research

Research Areas

History of Phonetics at The Ohio State University

Compiled by Ilse Lehiste, with the help of Mary Beckman, Robert A. Fox, and Keith Johnson.

1. Phonetics within the context of Speech Science

It is appropriate to start the history of phonetics at the Ohio State University in 1928 with the publica-tion of G. Oscar Russell’s epoch-making book, The Vowel, by The Ohio State University Press. Russell was a professor in the Romance Department, but he also offered some courses in phonetics. He was in-strumental in the establishment of the Phonetics Laboratory in 1929, and became its director.

The full title of Russell’s book is The Vowel: Its Physiological Mechanism as Shown by X-Ray, and in the first sentence of the introduction he states the purpose of the book as "to seek new light on the physiological causes of vowel quality differences". A re-reading of the volume shows his insights to have been far ahead of his time. He reviews the acoustic structure of the vowels as far as it was known, uses minimal pairs to establish auditory distinctions, and recommends using acoustic terms, which "stand for known facts wherever possible, in lieu of the now more or less current physiological expres-sions which the present study has apparently proved to be generally fantastic and practically without ba-sis in fact" (p.351). This emphasis on verifiable facts has characterized the phonetics program ever since.

Courses in experimental phonetics were offered under Russell’s supervision in the Romance Depart-ment until the establishment of the Department of Speech in 1936, with the Phonetics Laboratory as one of the components. The first doctorates were granted in 1936 and 1938. The program grew and at-tained national prominence under the leaadership of John W. Black, commencing in 1949. By 1955 the staff in Speech and Hearing consisted of eight faculty members and a laboratory technician.
In 1972, the Speech and Hearing area was elevated to Section status within the Department of Commu-nication. In 1988, the Section received Division status within the College of Social and Behavioral Sci-ences. In 1992, the Division became the Department of Speech and Hearing Science. In the academic year 1997/98, the department had a total faculty (including emeritus, adjunct and visiting professors) of 35.

The department has both a research component and a clinical component. Phonetics research is con-ducted under the leadership of Professors Lawrence L. Feth, Robert A. Fox, and Osamu Fujimura. Pro-fessor Feth has his doctorate from the University of Pittsburgh (1969) with a specialization in bioacous-tics. His research areas are auditory signal processing by normal and hearing impaired listeners, effects of noise on hearing, and models of speech perception. Linguistically oriented courses taught by Profes-sor Feth include Anatomy and Physiology of the Auditory System; Hearing Science; Physiological Acoustics; Psychoacoustics; and Psychoacoustics Seminar. Professor Fox has his Ph.D. from the Uni-versity of Chicago (1978) in Linguistics. He was associated with the Linguistics Department 1979-1983, teaching various courses in psycholinguistics. Moving to the Department of Speech and Hearing Science in 1983, he became Chairman of the Department in 1995. His current research deals with the interface between auditory and phonetic processing; rhythm in spoken language; structure of the mental lexicon; assessment of central auditory processing abilities, and age-related changes in speech percep-tion. Linguistically oriented courses taught by Professor Fox include Principles of Phonetics, Language Development, Speech Science, Articulation, Acoustic Phonetics, Experimental Phonetics, and Com-puter Use in Speech and Hearing. Professor Fujimura’s doctorate from the University of Tokyo (1962) is in the field of physics. His research areas are speech production and temporal organization, phonol-ogy and phonetics. He teaches an Advanced Seminar in Speech Science. The department offers sev-eral phonetics courses every year. It also has well-equipped laboratories for physiological and psycho-acoustic research in the areas of speech production and perception.

2. Phonetics in the Department of Linguistics.

The Department of Linguistics was established in 1965. From the very beginning, phonetics was an in-tegral component of the program. Phonetics was taught 1965-1985 by Professor Ilse Lehiste, who in 1965 became the first chairman of the new department after having spent two years in the Slavic De-partment. Professor Lehiste came to the Ohio State University from the University of Michigan, where she received her Ph.D. in linguistics in 1959, and spent 1959-63 at the Communication Sciences Labo-ratory as Research Associate. At Ohio State, she divided her time between phonetics, historical linguis-tics, and administration, serving as Chairman l965-7l, Acting Chairman l984-85, Chairman 1985-87, and Professor Emeritus since 1987.

As a rule, one general phonetics course and a seminar in advanced topics was offered every year. The seminar topics reflected Lehiste’s research interests, which were primarily focused on suprasegmental structure (a representative publication is the book Suprasegmentals, published in 1975 by MIT Press). Several extensive research projects were carried out, including the following. One of the areas concerned the phonetic realization of syntactic structure. The approach taken by Le-histe involved acoustic analysis of syntactically ambiguous sentences that speakers attempted to disam-biguate, accompanied by listening tests to establish the perceptual relevance of the acoustic cues that were discovered in the recordings. Representative publications include "The timing of utterances and linguistic boundaries", JASA 51, 6 (l972), 2018-2024, "Phonetic disambiguation of syntactic ambigu-ity", Glossa 7,2 (1973), 107-122, and "Rhythmic units and syntactic units in production and percep-tion", JASA 54, 5 (1973), 1228-1234.

The study of units of speech production and perception led to investigation of structural units larger than a sentence, especially the structure of paragraphs. The first such study involved a text consisting of three sentences that could be ordered in six possible ways ("The phonetic structure of paragraphs" in A. Cohen and S.G. Nooteboom, eds.: Structure and Process in Speech Perception: Proceedings of the Symposium on Dynamic Aspects of Speech Perception, Springer: Berlin – Heidelberg – New York (1975), 195-203.) The sentences were produced by a native speaker in isolation and as six paragraphs. A listening test was prepared in which sentences were presented in isolation. Thirty listeners took the test. The results showed that the listeners could tell whether the sentences had originally been produced in isolation or as part of the paragraph; they could even tell in which position within the paragraph the sentences had occurred. An acoustic analysis of the stimuli made it possible to isolate acoustic cues which the listeners may have used in making their judgments. The cues included different degrees of preboundary lengthening, and an intonation contour that declined systematically during the production of the paragraph.

The results were confirmed and extended in several studies, particularly in one using spontaneous speech (prompted monologue) as basis for a listening test ("Perception of sentence and paragraph boundaries", in B. Lindblom and S. Ohman, eds., Frontiers of Speech Communication Research, Aca-demic Press: London, New York, & San Francisco (1979), 191-202).

Throughout the period, Lehiste was engaged in the study of the prosody of two languages with compli-cated suprasegmental structures, Serbocroatian and Estonian. In collaboration with Pavle Ivic, a Ser-bian linguist, she investigated the accentual systems of standard Serbocroatian and three major dialects (Kajkavian, Slavonian, Cakavian). Over the course of 25 years, they published 16 articles and two books ( the second is a summary of all previous work: Ilse Lehiste and Pavle Ivic, Word and Sentence Prosody in Serbocroatian. (1986), MIT Press, Cambridge, Massachusetts, and London, England. The book was translated into Serbocroatian in 1996).

The investigation of the prosodic structure of Estonian focused on the phonetic realization of the quan-tity structure of Estonian, especially the system of ternary oppositions at the level of segment, syllable, and word. Here Lehiste used synthetic speech for controlled listening tests, but she had to produce the stimuli at other laboratories (Edinburgh, Bell Laboratories) , since at that time speech synthesis was not available at the Ohio State University. Her major finding was that the three-way opposition is realized primarily by contrastive duration ratios between the first (stressed) syllable and the following syllable.

The continued interest in prosody led to a major project that Lehiste is continuing after retirement: the phonetic investigation of metrical structure in orally produced poetry. The project deals with the rhyth-mic structure of poetry as compared to the rhythm of spoken language in general. A basic assumption in this project is the notion that the suprasegmental system of a language is crystallized in the metric struc-ture of its traditional poetry. Rhythm is part of the suprasegmental structure. To understand better the rhythm of spoken language, Lehiste proposes that one should look at the rhythmic structure of poetry developed in that language over the years; patterns that may be imperfectly realized in prose may be manifested in a more regular fashion in poetry. The rhythmic structure of poetry may just represent what for the realization of segmental sounds has sometimes been called "maximally differentiated style".

The project, which was supported by NSF, involves study of the phonetic realization of metrical struc-ture in orally produced poetry in the following languages: Finnish, Estonian, Latvian, Lithuanian, Ser-bocroatian, Swedish (both Finland-Swedish and Stockholm Swedish), Icelandic, and Faroese. The lan-guages represent both related and unrelated languages, traditionally classified as differing in their han-dling of rhythm. Another reason for the choice of these languages was Lehiste’s long-standing interest in language contact , especially in the North European geographic area ("Polytonicity in the area sur-rounding the Baltic Sea", in Nordic Prosody: Papers from a Symposium. Travaux de l’Institut de Pho-netique de Lund 13 (1978), 237-247). This project has led to numerous publications; among the find-ings is the establishment of the line as a unit of temporal programming, and the description of the pro-sodic structure of the stanza. Parallels were found to exist between the phonetic realization of metric foot, line, and stanza in poetry on the one hand, and foot, sentence, and paragraph in spoken language on the other hand. A representative publication is "The phonetics of metrics", Empirical Studies of the Arts 10,2 (1992), 95-120.

The study of the metric structure of poetry led further to a comparison between spoken and sung poetry. In her contribution to the Wenner-Gren International Symposium titled Music, Language, Speech, and Brain (Vol. 59 of the Wenner-Gren International Symposium Series, ed. by J. Sundberg, L. Nord and R. Carlson, London: Macmillan (1991), 98-107), she described the connection between speech and music as a series of gradual approximations with several intermediate stages, without abrupt boundaries be-tween them. The progression starts from spontaneous conversation and proceeds through read prose, rhythmic prose, free verse, poetry composed in one of the many regular metres, to orally transmitted poetry meant for recitation, poetry recited ‘by heart’, and poetry read from a provided text. The bridge toward music is built in several ways. Many cultures have a tradition of lamenting, which is half song, half recitation. Others have a way of presenting epic poetry in a monotonous chant. Lullabies can be sung as well as hummed, dances can be performed to a sung text or to nonsense syllables marking the beat. When the words are abstracted away and music is retained, one has arrived at the other end of this continuum.

In collaboration with a musicologist (Dr. Jaan Ross), Lehiste has analyzed poetic laments in Estonian dialects and in Karelian, and proceeded to analyze the manifestation of linguistically determined dur-ational patterns in folksongs (Jaan Ross and Ilse Lehiste, "Lost prosodic oppositions: A study of con-trastive duration in Estonian funeral laments", Language and Speech 37, 4 (1994), 407-424).

Research by graduate students during that time culminated in eight dissertations directed by Lehiste:
1970-1971: Zinny Bond, "Units in Speech Perception"

1973-1974: Linda R. Shockey, "Phonetic and Phonological Properties of Connected Speech" Sara S. Garnes, "Quantity in Icelandic: Production and Perception"

1976-1977: John Perkins, "An Acoustic Phonetic Study of Cross Dialect Borrowing" Holly Semiloff, "An Acoustic Correlate of Syllabicity in English" Robert K. Herbert, "Language Universals, Markedness Theory, and Natural Phonetic Processes: The Interaction of Nasal and Oral Consonants"

1981-1982: Christopher Farrar, "A Prototype Model of Speech Perception" Rachel Schaffer, "Vocal Cues for Irony in English"

1982-1983: Deborah Schaffer, "Intonation Cues to Management in Natural Conversation"

The dissertations of Bond and Shockey were published in the Working Papers in Linguistics series es-tablished by the department (of which fifty issues have been published l967-1997). The dissertation of Garnes was published in Hamburger Phonetische Beitrage Bd. 18, Hamburg: Buske (1976), and the dissertation of Herbert in Trends in Linguistics: Studies and Monographs 25, Berlin: Mouton de Gruyter (1986).

Mary R. Beckman joined the linguistics faculty in 1985. She has an A.B. degree in Oriental Lan-guages and Comparative Literature from the University of California, Berkeley, 1976, and an M.A. degree in Oriental Languages from the same university in 1979, followed by an M.A. degree in Lin-guistics from Cornell University in 1982, and a Ph.D. degree from Cornell in 1984. Beckman has ex-panded and diversified the phonetics offerings and has brought the laboratory up to date. She is co-founder (with John Kingston) of the Laboratory Phonology conference series and associated biannual collections of Papers in Laboratory Phonology. During the five years from 1990 through 1994, she also promoted laboratory approaches to fundamental issues in phonology as the editor of Journal of Phonetics (see, for example, the 1990 special issue on the theme of "Phonetic Representation"). Much of her own research focuses on prosody in all its aspects, from modeling the details of various phonetic correlates to developing computationally tractable phonological representations of stress and phrasing. She has published two monographs on aspects of prosody in English and Japanese (Stress and Non-Stress Accent, Netherlands Phonetic Archives #7, Foris Publications (1986), second printing Mouton de Gruyter (1992), and (with Janet Pierrehumbert) Japanese Tone Structure. Linguistic In-quiry Monograph Series 15, MIT Press (1988)) – and has developed several experimental paradigms for investigating articulatory and physiological representations of prosodic structure and timing. She has also done experimental work in first and second language phonological acquisition. (Some exam-ples of recent research topics: "Speech Models and Speech Synthesis", in J.P.H. van Santen, R. Sproat, J. Olive, and J.Hirschberg, eds., Progress in Speech Synthesis, Springer-Verlag (1977), 185-209, and "A typology of spontaneous speech", in Y. Sagisaka, W.N. Campbell, and N. Hi-guchi (eds.), Computing Prosody , Springer-Verlag (1997), 7-26.)

Beckman has directed the dissertations of the following students:
June 1988: Keith A. Johnson, "Processes of Speaker Normalization in Vowel Perception". Dr. Johnson was an Assistant Professor in the Department of Biocommunications at the University of Alabama in Birmingham before taking up his current position in the Linguistics Department at Ohio State Univer-sity in 1993.
August 1991: Kenneth J. de Jong, "The Oral Articulation of English Stress Accent". Dr. de Jong is an Assistant Professor in the Department of Linguistics of Indiana University.
August 1993: Sun-Ah Jun, "The Phonetics and Phonology of Korean Prosody". Dr. Jun is an Assistant Professor in the Department of Linguistics of the University of California at Los Angeles.
January 1994: Sook-hyang Lee, "The Role of the Jaw in Consonant Articulation". Dr. Lee is an Assis-tant Professor in the Department of English Linguistics and Literature, Wonkwang University, Chun-buk, Korea.
January 1996: Shu-hui Peng, "Phonetic Implementation and Perception of Segmental Coarticulation and Tone Sandhi". Dr. Peng is currently a visiting lecturer in the Department of Linguistics, Univer-sity of Alberta, Edmonton, Canada.
January 1996: Gayle Ayers, "Nuclear Accent Types and Prominence: Some Psycholinguistic Experi-ments". Dr. Ayers is a member of the technical staff at Entropics, Inc., Washington, D.C., a leading provider of software for speech analysis. A recently completed project was the development of an en-hanced tool for prosodic transcription of speech databases using the ToBI system for English.

Keith A. Johnson joined the department in 1993. He received his B.A. degree at Abilene Christian University in 1981, his M.A. degree in Linguistics at the Ohio State University in 1985, and his Ph.D. degree at the same university in 1988. His area of specialization is acoustic and articulatory phonetics, and the phonetics/phonology interface. Current projects include impact of gender differences in speech perception; individual differences in speech production; phonetic characterization of Native American languages; and phonetics of ventriloquism. He is the author of Acoustic and Auditory Phonetics , Cambridge: Blackwell (1977), and editor (with John W. Mullennix) of Talker Variability in Speech Processing, San Diego: Academic Press (1997). Representative articles include "Contrast and nor-malization in vowel perception", Journal of Phonetics 18 (1990), 229-254; "Differential effects of speaker and vowel variability on fricative perception", Language and Speech 34 (1991), 265-279, and (with J.V.Ralston) "Automaticity in speech perception: Some speech/nonspeech comparisons", Pho-netica 51 (1994), 195-209. Keith Johnson moved to the University of California, Berkeley in 2004.

Cynthia Clopper joined the department in 2006. She received an A.B. in Linguistics and Russian from Duke University in 1999, an M.A. in Linguistics from Indiana University in 2001, and a PhD in Linguistics and Cognitive Science from Indiana in 2004. Clopper’s research interests include speech perception, sociophonetics, and laboratory phonology. Her current research projects include the inter-action between dialect variation and spoken language processing, an analysis of prosodic sources of regional dialect variation in the United States, and implicit accommodation to and explicit imitation of regional dialects. Representative articles include "Acoustic characteristics of the vowel systems of six regional varieties of American English", Journal of the Acoustical Society of America 118 (2005), 1661-1676 (with D. B. Pisoni and K. de Jong); "Perceptual similarity of regional varieties of American English," Journal of the Acoustical Society of America 119 (2006), 566-574 (with S. V. Levi and D. B. Pisoni); and "Free classification of regional dialects of American English", Journal of Phonetics 35 (2007), 421-438 (with D. B. Pisoni).

The phonetics courses offered by the Department of Linguistics include the following.
Ling. 500, Phonetics. Cross-linguistic survey of the sounds of the world’s languages.
Ling. 600.01, Phonetic Theory. Principles of articulatory phonetics, with some discussion of acoustic phonetics; practice in the production, recognition, and transcription of sounds in various languages of the world. In this five-credit course, there will be three hours of lectures, one hour of demonstration, and one review session per week. The weekly demonstration will be either an in-class language tran-scription practice from live speech or a laboratory session illustrating techniques for gathering data in speech acoustics, perception, or articulation. Coursework consists of a small field project, midterm and final exams, and weekly homework (including transcription exercises from tape as well as prob-lems and essay questions). Designed to introduce students to the principles and mechanisms of speech production and perception, Ling. 600 serves a threefold purpose in our curriculum. First, it provides the articulatory, acoustic, and perceptual framework for the discussion of phonetic processes, thus serving as a background to our courses in phonology. Second, the attention given to production, recog-nition, and transcription of a wide variety of sounds of different languages serves to develop expertise applicable in linguistic field work. Finally, the coverage of basic elements of various areas of phonet-ics is aimed at providing the necessary background for further work in phonetics, such as the interme-diate laboratory course (Ling. 600.02).
Ling. 600.02, Laboratory Phonology. Introduction to laboratory methods and quantitative models of speech for linguistics.This course provides an intermediate-level course between the introductory course in phonetics (Ling. 600.01) and the seminar in advanced phonetics (Ling. 825). It is also an in-troduction to methods of experimental research and computational modeling of speech perception and production as tools in phonological analysis. There will be two two-hour sessions each week devoted either to lecture and discussion in the classroom or to demonstration of particular techniques and data in the Linguistics Laboratory. The major coursework will consist of weekly problem sets involving the gathering and/or analysis of real speech data in published literature exemplifying the laboratory pho-nology approach.
Ling. 825, Seminar in Advanced Phonetics. Study of specific problems in articulatory and acoustic phonetics at an advanced level. Students will be expected to read the assigned papers, participate in discussions and assigned class projects, and then carry out a final project related to the subject matter of the seminar together with a write-up of the project.