Acoustical Society of America
Gold Medal Award - 1995

Kenneth N. Stevens

KEN STEVENS has been a member of this society for 45 years. During that period of time he has been a faithful and durable jogger, running, we estimate, some 13,500 miles, or the distance from here to Sydney, Australia. And this is merely one of his memorable feats. When Ken hasn't been running, cycling, playing the harpsichord, or tinkering with his cottage in Liberty, Maine, he has been distinguishing himself as our foremost student of speech.

Ken was born in Toronto, Canada in 1924 and received his Bachelor's and Master's degrees in Electrical Engineering from the University of Toronto, and an Sc.D. from the Massachusetts Institute of Technology (MIT). Ken's doctoral thesis, "The Perception of Sound Shaped by Resonant Circuits," was supervised by Leo Beranek who, in 1975, was himself awarded the Gold Medal of the Acoustical Society of America. Ken learned well from his mentor.

In 1951 Ken joined the faculty of MIT. He became a full professor in 1963, and currently holds the Lebel chair in the Department of Electrical Engineering and Computer Science. He is the head of the Speech Group, a research team of eight faculty and staff plus about nine graduate students and post-doctoral fellows who are housed in the Research Laboratory of Electronics. Commenting on Ken's unique style of managing a research team, one of Ken's closest associates, the late Dennis Klatt, wrote "As a leader, Ken is known for his devotion to students and his miraculous ability to run a busy laboratory while appearing to manage by a principle of benevolent anarchy." To this well-deserved praise one might add that during its existence of almost four decades Stevens' Speech Group at MIT has been outstanding in the support that it has provided to women researchers, many of whom have gone on to populate the upper echelons of research labs throughout the world.

One of Ken's major contributions concerns the structure of speech sounds. Thanks to his work we have begun to appreciate the intricate manner in which different aspects of the human anatomy are exploited in the production of speech. We speak at the rate of 180 words a minute or 15 speech sounds per second. To get a realistic grasp of this spectacular accomplishment, consider that each speech sound is chosen from an alphabet that, in the case of English, consists of 40 sounds. Since each speech sound has its own configuration of articulators—i.e., of the tongue, tips, velum, and larynx—we choose from 40 different articulator configurations 15 times per second whenever we say something in English. How even the clumsiest person can move the tongue, lips, velum, and larynx fast enough to accomplish this is one of the classical puzzles of speech science, and we owe to Ken's research an essential part of the answer to this conundrum.

Ken noticed that the relationship between articulator movement and the acoustic signal that this produces is, in many cases, nonlinear. For instance, both /s/ and /sh/ are produced by touching the front of the hard palate with the front part of the tongue. When contact is more anterior, we get /s/. When contact is made slightly further back, we get /sh/. As one retracts the tongue from the /s/ position to the /sh/ position, the acoustic output remains unchanged for a considerable part of the retraction.

However, once one passes a certain point, roughly the alveolar ridge, the output changes abruptly to /sh/ and remains /sh/ from then on. Thus, in order to distinguish /s/ and /sh/ reliably, one needs only place the tongue in front of or behind the alveolar ridge and the correct acoustic output is always produced without the need for any special accuracy.

In his work on the quantal nature of speech, Ken has attempted to show that many distinctive features—distinctions between speech sounds—are of this kind; i.e., they utilize special nonlinear relations between articulation and acoustic output that enable speakers to produce the correct sound without having to hit all of the individual targets with particular accuracy. And because of the low-accuracy requirements, even the clumsiest among us can produce 15 sounds per second. In this way Ken has unraveled an important part of the mystery that shrouds our ability to produce and to understand speech. As part of this work, Ken has also made major contributions to our understanding of the physics and aerodynamics of sound source generation.

On the perceptual side, it has been difficult to understand how, in the face of variation induced in the sounds of words by context, casualness and noise in the environment, listeners can still understand what speakers have said. Ken's work suggests strongly that, in spite of this variation, enough of the right kind of acoustic detail (i.e., feature correlates) remains in the signal to account for this ability.

Ken was involved in the earliest work on speech synthesis, modeling the time-varying shape of the vocal tract. The name of the device that he helped to create was called DAVO, an acronym for Dynamic Analog of the Vocal Tract. As it happened, much of this work was taking place during the height of the Cold War. Those of you who have read Solzhenitsyn's The First Circle will recall the group of scientists working on much the same problem in the Mavrino Prison, outside Moscow. Despite the intensity of the Cold War, Ken received a number of Russian scientists during this period. In his legendary hospitable fashion, Ken introduced them to DAVO. On one occasion he put hospitality to one side long enough to ask one of his visitors. "Why in the world are you working on this stuff?" "Because, his visitor answered, "You are."

Throughout his academic life Ken has received a number of honors, including the Lebel chair mentioned above and the Silver Medal in Speech Communication from his Society. He is being honored now with the Society's highest award, the Gold Medal of the Acoustical Society of America. This is a fitting culmination to a career which has gained for Ken the affection and respect of his colleagues and students in response to his own devotion to them over the course of a remarkable scholarly life. Morris Halle
Samuel Jay Keyser
Joseph Perkell
Stefanie Shattuck-Hufnagel