Voice-stress software is put to the test
(Phys.org) -- Computer scientists working in a range of departments, from pervasive computing to mobile computing to sensors, will converge in Pittsburgh for the 14th International Conference on Ubiquitous Computing (Ubicomp 2012) on September 5 through 8. They will learn more there about a teams software that can recognize stress from vocal patterns. The researchers wanted to see if they could successfully train a system to pick up stress through a person's voice. Their paper, StressSense: Detecting Stress in Unconstrained Acoustic Environments using Smartphones, authored by Hong Lu, Mashfiqui Rabbi, Gokul T. Chittaranjan, Denise Frauendorfer, Marianne Schmid Mast, Andrew T. Campbell, Daniel Gatica-Perez, and Tanzeem Choudhury, details their success.