SBIR Phase II: Automated Public Speaking Assessment

This Phase II project aims to develop software to automatically assess public speaking skills and prepare students with better oral communications skills necessary to perform job tasks. Oral expression is the most highly valued ability throughout the economy and ranks as the second most highly-valued skill for high-wage, high-growth, high-skill occupations. Approximately 4.5 million college students take a basic communications course each year, however, as class sizes get larger and online learning becomes more common, public speaking instruction becomes increasingly difficult. Practice and feedback are essential aspects of these courses, yet it is a struggle for teachers to find enough time to sufficiently interact with students. This SBIR project aims to develop the key concepts of automated public speaking assessment such that a student?s vocal delivery can be objectively measured and presented in a manner that creates an independent, personalized learning experience. Unlike traditional methods of public speaking assessment, the proposed system can be available at any time, provide objective feedback and track student practice and improvement. The proposed Software-as-a-Service is projected to generate $16 Million in revenue over five years and create more than 25 high-paying, US-based jobs. This Small Business Innovative Research (SBIR) Phase II project proposes to develop an automated assessment system for public speaking that determines how a speaker would be perceived by an audience. Automated assessment for speech has already occurred in spoken language proficiency, which leverages Automated Speech Recognition (ASR) and semantic analysis. Automated voice assessment has also been utilized in lie detection and emotion detection, which focus on autonomic responses in the user?s voice, such as when stress affects the vocal cords. The hypothesis behind this SBIR project is that speakers can consciously use and modify non-semantic speech behaviors to produce more desirable listener perceptions. Automatically linking listener perception to speech behaviors represents a novel direction in automated assessment for speech. The Phase II objective is to develop software sufficient for automated public speaking assessment such that a student?s vocal delivery can be objectively measured and presented in a manner that creates an independent, personalized learning experience. Voice analytics capability investigated in Phase I will be enhanced and developed into a cloud-based service which helps students practice, track, and improve their public speaking habits.