How does your tongue move when you talk? Scientists have long known that minute differences in tongue configuration may produce significantly different sounds. Yet study of tongue movement faces a significant challenge - the tongue is hidden inside the mouth. The National Science Foundation will provide partial support for the third Ultrafest workshop to be held at the University of Arizona. Recent advances in ultrasound technology suggest it as more practical and user-friendly than the familiar alternatives (like x-rays, palatography, magnetic resonance imaging, etc.): It is noninvasive, nontoxic, relatively inexpensive (once the initial machine cost is covered), and produces real-time video images. Ultrasound study oriented towards understanding tongue movement during normal and abnormal speech is in its infancy. This workshop brings together the world's scientists who are using ultrasound to study language articulation, to share descriptions of each lab's research foci, to explore methodological and data analysis issues, and to present results of current research projects. The format is that of a workshop rather than a more formal conference, with faculty, post docs, and students sitting round the table together sharing ideas. Critical methodological issues include (a) rate of data extraction (speeding up both extraction of frames from the video and identification of data points showing the edge of the tongue in each frame); (b) image adjustment to eliminate effects of subject movement during the recording session; (c) image improvement to enhance the quality of the tongue profile and to add the palate in the appropriate place to the tongue image; (d) analytic challenges (for each different type of research question, what should be measured and how should statistics be calculated?).

Ultrasound in language research is an innovative application of an existing technology. As a phonetic tool, ultrasound can bring about a veritable explosion of new data with which to revisit old research questions and pose new ones. To achieve this, as with any new research methodology, it is critical to develop and test standards to improve replicability and reliability of results. The workshop approach is designed for maximum open interaction among the participants geared to exploring and improving solutions to shared problems.

Project Start
Project End
Budget Start
2005-01-01
Budget End
2006-06-30
Support Year
Fiscal Year
2004
Total Cost
$16,718
Indirect Cost
Name
University of Arizona
Department
Type
DUNS #
City
Tucson
State
AZ
Country
United States
Zip Code
85721