Page 50 - 2016Winter

Maureen Stone
Postal:
Department of Neural and Pain Sciences and Department of Orthodontics University of Maryland School of Dentistry 650 West Baltimore Street Baltimore, Maryland 21201 USA
Email:
stone@umaryland.edu
Christine H. Shadle
Postal
Haskins Laboratories 300 George Street New Haven, Connecticut 06511 USA
Email:
shadle@haskins.yale.edu
A History of Speech Production Research
Many of the initial assumptions about speech and how we produce it turned out to be wrong; these myths reveal the complexity of the process.
Introduction
Researchers have studied speech production, which is everything that goes into talking from the thought to the articulator movements to the sound emerging from the mouth, to understand it, transmit it over long distances, mimic it, and treat speech pathologies. Progress has not been steady. Sometimes researchers had to trip over their own wrong assumptions before they could take the next step.
We think of “speech” as an acoustic signal that is heard when people talk to each other. But that sound is the end result of a complex process, starting in the brain where thoughts are translated into language and muscle commands are sent to our vocal tract to shape it and get air moving through it. As the air moves through the slowly changing vocal tract, the right sounds will be produced in the right se- quence so that they can be interpreted by the listener as speech. Once the muscles are activated, production is in progress and the physical properties of aerodynam- ics take over to produce a particular acoustic waveform. Given the variety of vocal tract shapes possible, predicting the sounds that will be produced is itself complex.
Speech production includes all the processes involved in producing the acoustic waveform described as speech. Speech perception includes all the processes in- volved in receiving, processing, and interpreting that waveform as speech. In this article, we describe some of the pivotal points that have changed our understand- ing of speech production over the years.
Milestones in science often come from radical changes in our understanding of how things work. In speech production, a number of such changes were crucial to the development of the field. Therefore, we have organized this article as a series of “myths,” that were well believed until advances in the sophistication of our instru- mentation, the richness of our data, and our knowledge of other fields inspired advancements in our thinking.
Myth 1: Speech Is Made of Concatenated Sounds with Silence
Between Words
In 1877, Henry Sweet, a source for Pygmalion’s Henry Higgins, developed the sci- ence of phonetics and used it to describe Received Pronunciation, a dialect of British English traditionally associated with the nobility. At the same time, Alex- ander Melville Bell (father of Alexander Graham Bell) developed visible speech (Bell, 1867), which is a set of graphic diagrams of articulatory positions used to teach the deaf to learn to speak. These were early efforts on the part of linguists and speech pathologists to use an orthographic alphabet to capture features of concatenated oral sounds.
In the 1930s, interest in speech disorders grew. In 1937, Charles Van Riper, a stutterer himself, developed a scientific basis for the research and remediation of stuttering. The 2010 film The King’s Speech presents the application of such tech-
48 | Acoustics Today | Winter 2016 | volume 12, issue 4 ©2016 Acoustical Society of America. All rights reserved.