TY - JOUR
T1 - Interactive virtual reality speech simulation system using autonomous audience with natural non-verbal behavior
AU - Liao, Justin Andrew
AU - Jincho, Nobuyuki
AU - Kikuchi, Hideaki
N1 - Funding Information:
Manuscript received May 2, 2018; revised July 16, 2018. This work was supported by the Japan Society for the Promotion of Science KAKENHI under Grant Number 17KT0141.
Publisher Copyright:
© 2018, International Association of Computer Science and Information Technology.
PY - 2018/8/1
Y1 - 2018/8/1
N2 - Public speaking anxiety (PSA) is a fear of speaking in front of others. Most people experience a certain amount of anxiety in public speaking situation. This study aims to help people overcome PSA using an interactive VR simulation system with real-life scenarios. We present a multimodal VR speech simulation system using autonomous audience with natural non-verbal behavior to enhance users' sense of presence. Additionally, real-time multimodal feedback is produced by virtual audience based on users' public speaking behavior which automatically analyzed by multimodal sensors (e.g. microphone, motion capture, heart rate monitor). We perform an evaluation based on self-assessment questionnaires and biometry to investigate three study conditions: (I) control condition (baseline), (II) interactive virtual audience, and (III) virtual audience with natural non-verbal behavior. We divided participants into two groups with different conditions: interactive virtual audience condition (n = 7) and virtual audience with nature non-verbal behavior condition (n = 9). The results indicate that the usage of a virtual audience with natural non-verbal behavior increased a higher sense of presence and more anxiety-provoking.
AB - Public speaking anxiety (PSA) is a fear of speaking in front of others. Most people experience a certain amount of anxiety in public speaking situation. This study aims to help people overcome PSA using an interactive VR simulation system with real-life scenarios. We present a multimodal VR speech simulation system using autonomous audience with natural non-verbal behavior to enhance users' sense of presence. Additionally, real-time multimodal feedback is produced by virtual audience based on users' public speaking behavior which automatically analyzed by multimodal sensors (e.g. microphone, motion capture, heart rate monitor). We perform an evaluation based on self-assessment questionnaires and biometry to investigate three study conditions: (I) control condition (baseline), (II) interactive virtual audience, and (III) virtual audience with natural non-verbal behavior. We divided participants into two groups with different conditions: interactive virtual audience condition (n = 7) and virtual audience with nature non-verbal behavior condition (n = 9). The results indicate that the usage of a virtual audience with natural non-verbal behavior increased a higher sense of presence and more anxiety-provoking.
KW - Autonomous virtual audience
KW - Non-verbal behavior
KW - Public speaking anxiety
KW - Virtual reality
UR - http://www.scopus.com/inward/record.url?scp=85051846119&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85051846119&partnerID=8YFLogxK
U2 - 10.18178/ijmlc.2018.8.4.720
DO - 10.18178/ijmlc.2018.8.4.720
M3 - Article
AN - SCOPUS:85051846119
SN - 2010-3700
VL - 8
SP - 404
EP - 407
JO - International Journal of Machine Learning and Computing
JF - International Journal of Machine Learning and Computing
IS - 4
ER -