From the left, Shri Narayanan, Rahul Gupta, Kartik Audhkhasi, Sungbok Lee
A team of USC researchers from the Signal Analysis and Interpretation Laboratory (SAIL), directed by Professor Shri Narayanan, of the Ming Hsieh Department of Electrical Engineering won the INTERSPEECH 2013 Computational Paralinguistics Challenge Award. The team was led by EE PhD students Rahul Gupta and Kartik Audhkhasi, and included Professors Sungbok Lee and Shrikanth Narayanan. This is a record fourth such award for USC SAIL: teams from USC SAIL were winners of the INTERSPEECH challenge awards in 2009, 2011 and 2012. INTERSPEECH is the world's largest interdisciplinary conference focused on speech and language science and technology. [http://www.interspeech2013.org/]
The award recognizes the research paper describing an approach that achieved the best performance in an international competition for automated recognition of vocally-expressed paralinguistic signals. The paper was presented at the 14th Annual Conference of the International Speech Communication Association held August 2013. The signal processing and pattern recognition algorithms of the competitors were tested on a database supplied by the organizers. The winning paper of USC SAIL is entitled “Paralinguistic event detection from speech using probabilistic time-series smoothing and masking”.
Their approach relied on modeling temporal context surrounding paralinguistic events in speech such as laughter and fillers like ‘um’. Additionally novel probabilistic masking techniques to remove false alarms were proposed. The ability to analyze and model such paralinguistic events offers insights into cognitive underpinnings but also affective and social aspects of behavior. Details of this research, and on going efforts on human behavioral signal processing and behavioral informatics with applications to domains ranging from commerce and security to health realms such as Autism and addiction, can be found at http://sail.usc.edu