Introducing WASL: A new open-source MATLAB Toolbox for Multimodal Speech Production Research

Speaker: Dr. Steven M. Lulich
Affiliation: Indiana University, Bloomington

Abstract:

WASL is a new open-source MATLAB toolbox for multimodal speech research, developed in the Speech Production Laboratory at Indiana University. The toolbox is extensible through the development of plug-ins, and several plug-ins are already available. Basic and extended WASL functionality will be demonstrated through an analysis of real-time volumetric ultrasound images of American English rhotic and lateral sounds produced by adults and 1st-grade children. An anatomically-oriented Principal Components Analysis (PCA) anchored at the mandibular symphysis (where the tendon of the genioglossus muscle attaches to the mandible) will presented as a means for quantifying tongue surface shapes, and its relation to prior 2D and 3D PCA analyses of tongue shapes will be discussed.

Biography: 

Dr. Lulich earned a BA in Linguistics from Dartmouth College in 2002 and a PhD in Speech & Hearing Bioscience and Technology from MIT in 2006. From 2006 to 2013, Dr. Lulich’s research focused primarily on acoustic modeling and analysis of speech, including the subglottal airways. He joined the faculty in the Department of Speech & Hearing Sciences at Indiana University in 2013, where his current research focuses on modeling and analysis of speech production, especially the three-dimensional tongue shape and its variability during speech using real-time 3D/4D ultrasound.

For more information, contact Abeer Alwan (alwan@ucla.ed)

Date/Time:
Date(s) - Feb 17, 2017
12:00 pm - 2:00 pm

Location:
E-IV Tesla Room #53-125
420 Westwood Plaza - 5th Flr., Los Angeles CA 90095