Acoustical Society of America
R. Bruce Lindsay Award - 2003

Dani Byrd

Dani Byrd is a phonetician whose accounts of the movements of the vocal organs are literally dynamic. She has described the gestures we make with our tongues and lips in ways that reveal many new facets of their organization. Her career in acoustics began in 1986 when she left the warmth of southern California to attend Yale University. There, through Louis Goldstein of the Yale Linguistics Department and Haskins Laboratories, she discovered phonetics and speech science, and earned a joint BA/MA in Linguistics. At Yale she also competed in fencing; her future husband, Oliver Foellmer, was another new member of the Yale team. Dani was described to us by a fencing colleague as intense, a quality that characterizes her more generally, in all her pursuits.

We have known her since she began the Ph.D. program in the Phonetics Lab of the Linguistics Department at the University of California, Los Angeles (UCLA) in 1990, Peter Ladefoged serving as adviser for her MA equivalency, and Pat Keating as her dissertation chair. Dani, who is hardworking and ambitious, hit the ground running at UCLA, quickly turning her Yale MA thesis into her first publication. Since then she has not slowed down. Her dissertation set her on her long-term research program to understand how linguistic structure conditions the temporal realization of speech. Phoneticians have not fully understood how the units of speech are joined together to produce an acoustic signal, and Dani has been in the forefront of researchers tackling this fundamental issue. She has been using many tools to study speech production, including instruments that measure contact of the tongue on the palate, and positions of fleshpoints tracked in an electromagnetic field. In her dissertation she studied the spatial and temporal characteristics (that is, the articulatory constrictions, the durations, and the overlap) of English consonants under a variety of linguistic conditions. The major contribution of this work has been the idea that these characteristics do vary systematically as a function of linguistic structure; in particular, that different articulations are more strictly timed in smaller linguistic units. She proposed a model of multiple influences on articulatory timing, which she has since been developing in collaboration with Elliot Saltzman of the Boston University Department of Physical Therapy and Haskins Labs.

Throughout her time at UCLA Dani was involved in a variety of other projects. Of these, probably the most important was the work on phonetic analysis of the Texas Instruments—Massachusetts Institute of Technology (TIMIT) database of American English speech, the preliminary results of which were published in the Journal in 1992. She showed that a number of phonetic variables indexing segmental reduction distinguish the men from the women. This is work with not only theoretical implications, but also practical value for those working on speech synthesis and recognition. In addition to setting a standard with her many research projects, she was an outstanding colleague in the Phonetics Lab and the Linguistics Department; she was simply one of those students who helped everyone and was relied on by everyone. Her achievements as a graduate student at UCLA were recognized by two awards: a College award, and a Graduate Woman of the Year award.

Upon finishing her UCLA degree, she married Oliver, returned to Haskins Laboratories as a National Institutes of Health (NIH) postdoctoral trainee, and began projects with not only Elliott Saltzman, but also several others. She collaborated with engineers and linguists using structural MRI to determine the articulation of some unusual consonants of Tamil (a study published in the Journal; with Goldstein and others in a methodological comparison of electromagnetic articulography and the X-ray microbeam; with the Yale fMRI group; and with Katherine Harris on apraxic speech production. After her traineeship, she remained at Haskins as a staff researcher, funded by an NIH FIRST award. The major contribution of the research supported by this grant and by her continuing collaboration with Saltzman has been their model of "prosodic gestures"—speech articulations that are not tied to particular consonants, vowels, or articulators, but which affect any speech gestures in a given linguistic structure. This is a completely novel approach to how linguistic structure could directly affect articulation. Dani has taken the equations describing articulatory motion that Saltzman and others have developed, and shown that they can be practically applied to real linguistic data in a way that helps us understand the nature of speech as a manifestation of language. She is enlightening us about both the theoretical concepts underlying articulatory motion, and how those concepts can be related to linguistic analysis.

In Fall 1999 Dani was able to leave cold weather behind forever when she began as an Assistant Professor of Linguistics at the University of Southern California (USC), where she has recently been promoted to Associate Professor. As Joseph Aoun, Dean of the College of Letters, Arts, and Sciences at USC, says, "Dani Byrd single-handedly built a phonetics laboratory and a phonetics program and added substantially to the phonology program at USC. Her research is highly interdisciplinary and attracts distinguished faculty and students from the Department of Psychology and the School of Engineering." Among these collaborations have been her recent projects with Shri Narayanan of the USC Electrical Engineering Department and others on realtime magnetic resonance imaging of speech production, and on spoken language interaction between preschoolers and computers. Dani's experience as a mother of a young daughter no doubt informs this latter project,"Politeness and frustration language in child-machine interactions."

As even a glance at her bibliography will show. Dani personifies the interdisciplinary character of the Speech Communication Technical Committee, and the Society as a whole. She has been active with professional service, including serving on a National Science Foundation panel. Her involvement in the Society includes, of course, attending many of the meetings (beginning in 1990), publishing in the Journal, and membership on the Speech Communication TC, including serving on the sub-committee charged with selecting the recipient of the Raymond H. Stetson Scholarship, and as co-organizer of the Speech Communication program at the Newport Beach meeting. We anticipate that she will serve the Society in many ways in the future.

PATRICA A. KEATING

PETER LADEFORGED