INCOSE Orlando Chapter Meeting: Automatic Speech Recognition in Training Systems: Misconceptions, Challenges and Paths Forward

Jan 21, 2016 5:45 PM - Jan 21, 2016 7:15 PM Eastern Standard Time

Abstract:
Over the past decade, we have seen moderate demand for simulation-based training systems to include automatic speech recognition (ASR).  Similar to commercially available ASR services such as Apple's Siri and Google Now, this capability would allow training systems the ability to interpret human speech and react to that speech with appropriate actions (e.g. executing a spoken command) and responses (e.g. replying to a human with confirmation or requests for clarification).  However, ASR successes within simulation-based training systems have been moderate at best.  We contend that this lack of widespread usage and success stems first and foremost from a fundamental misunderstanding of (and thus lack of investment in) the components necessary to achieve effective ASR. 

In this talk, Brian will introduce the various functions of ASR (recognition, understanding, execution and dialog), how they are typically implemented, and where trouble points lie. We will conclude with a brief description of the current state of the art and a discussion of the roadblocks and opportunities for bringing the state-of-the-art can be applied to contemporary simulated training systems.

Presenter:
Dr. Brian Stensrud, Soar Technology Inc., Senior Scientist

Biography:
Dr. Brian Stensrud is a Senior Scientist at Soar Technology, Inc. (SoarTech) in Orlando, FL. Brian received a Ph.D. in Computer Engineering from the University of Central Florida (2005), and BS degrees in Mathematics and Electrical Engineering from the University of Florida (2001). He has 15 years experience in artificial intelligence research and development and his professional interests include ASR and dialog capabilities for training simulations, automatic population of knowledge-rich behavior models, and intelligent agents for virtual training and practice environments.