A Multimodal Corpus for the Assessment of Public Speaking Ability and Anxiety

A Multimodal Corpus for the Assessment of Public Speaking Ability and Anxiety Mathieu Chollet1, Torsten W¨ortwein2, Louis-Philippe Morency3, Stefan Scherer1 1 Institute for Creative Technologies, University of Southern California, Los Angeles, CA, USA 2 Institute of Anthropomatics and Robotics, Karlsruhe Institute of Technology, Karlsruhe, Germany 3 Language Technologies Institute, Carnegie Mellon University, Pittsburgh, PA, USA {mchollet, scherer}@ict.usc.edu, uncwt@student.kit.edu, morency@cs.cmu.edu Abstract The ability to efficiently speak in public is an essential asset for many professions and is used in everyday life. As such, tools enabling the improvement of public speaking performance and the assessment and mitigation of anxiety related to public speaking would be very useful. Multimodal interaction technologies, such as computer vision and embodied conversational agents, have recently been investigated for the training and assessment of interpersonal skills. Once central requirement for these technologies is multimodal corpora for training machine learning models. This paper addresses the need of these technologies by presenting and sharing a multimodal corpus of public speaking presentations. These presentations were collected in an experimental study investigating the potential of interactive virtual audiences for public speaking training. This corpus includes audio-visual data and automatically extracted features, measures of public speaking anxiety and personality, annotations of participants’ behaviors and expert ratings of behavioral aspects and overall performance of the presenters. We hope this corpus will help other research teams in developing tools for supporting public speaking training. Keywords: Multimodal corpus; public speaking training; virtual audiences 1. Introduction Modern life often involves situations where we are required to speak in public, both in our professional lives, for in- stance when presenting results of our work in front of col- leagues, and in our personal lives, such as when giving a toast at a wedding. A proficient public speaker mobilizes a vast array of skills to perform a good speech, ranging from the selection and arrangement of appropriate and convinc- ing arguments to the efficient vocal and non-verbal delivery of the speech. Considering how prevalent public speaking situations are in modern professional and personal life, it is only natural some individuals would desire to improve their ability to speak in public. Additionally, public speak- ing anxiety is an extremely common fear (Furmark et al., 2000; Bodie, 2010), and some people experience an un- manageable amount of stress when preparing or undergoing public speaking. These two situations warrant for the de- velopment of tools and methods to support the assessment of public speaking ability, the training of public speaking skills and the reduction of public speaking anxiety. Multimodal interaction technologies (e.g. social signals processing, virtual humans) have been deployed in many types of social skills training applications, from job inter- view training (Hoque et al., 2013; Ben Youssef et al., 2015) to intercultural skills training (Lane et al., 2013) or public speaking skills training (Damian et al., 2015; Chollet et al., 2015). Moreover, virtual audiences have been used for sup- porting people suffering from severe public speaking anx- iety (North et al., 1998; Pertaub et al., 2002). Finally, re- cent works …

Content truncated to 500 words