Our Methods
To investigate speech production, the Speech Kinematics and Acoustics Lab at Vanderbilt uses motion capture technology to quantify speech motor performance. The lab is equipped with a six-camera infrared motion capture technology (Motion Analysis, Inc.) and a 3D electromagnetic articulograph (AG501; Carstens Medizinelektronik, Germany).
Infra-red cameras are used to track facial motion during speech. The reflective stickers are placed around the mouth and on the jaw, reference stickers are placed on the forehead and nose. In Hollywood the same technology is used to create animation movies such as “The Polar Express”. In our lab, we use the motion capture technology to better understand speech motor performance deficits of speakers with brain diseases, detect subtle changes in motor performance due to medication or behavioral treatments, and track disease progression of degenerative diseases.
Electromagnetic articulography can be used to record movements of the tongue during speech while also recording lip and jaw movements. Sensors are attached to tongue (tongue tip, tongue body), lower teeth (for jaw movement), upper and lower lips, and head (for head movement tracking - not shown in this figure). Sensors move with the articulators as the speaker produces speech. These sensor movements can be seen in 3D in real time during the data collection. As can be seen, different structures produce different movement patterns – the lower lip and the tongue, for example, can produce relatively large and relatively fast speech movements, whereas the jaw and the upper lip move much slower and to a lesser extent than lower lip and tongue. One working hypothesis that we are currently investigating in the lab is that speakers may “trade” the use of speech structures that are impacted by a disease for those that are less impacted. We analyze further how such trade-offs impact the speech acoustic signal.
