Huang's paper, "Automated Speech Recognition Technology to Support in Flight Weather-Related Communication for GA Pilots", explores speech recognition interfaces to support General Aviation pilots. The work is part of the Federal Aviation Administration's PEGASAS (Partnership to Enhance General Aviation Safety, Accessibility and Sustainability) Center of Excellence. Huang was also a finalist in The Stanley Nelson Roscoe Best Student Paper Competition. 

The ISAP conference was held in Dayton, OH, May 7-10. ISAP focuses bringing together researchers who are interested in human performance problems and opportunities within aviation systems. 

ABSTRACT

Weather information latency during flight in general aviation (GA) has resulted in numerous incidents. Hands-free automated speech recognition (ASR) systems have the potential to help overcome this challenge and facilitate rapid weather-related information exchange. However, it is unclear to what extent ASR systems can support pilot communication in such noisy environments. The goals of this study were to (1) evaluate the performance of 7 commercially-available ASR systems to recognize weather phrases during GA operations and (2) determine whether speech-to-noise (S/N) ratio, flight phase, and accent type modulate system performance. Overall, the highest accuracy percentage achieved by any system was 72%, when the S/N ratio was at least 3/2. This research can help to inform the selection and development of next-generation technologies to be used in safety-critical, information-rich domains.