TY - GEN

T1 - Prediction of Audience Response from Spoken Sequences, Speech Pauses and Co-speech Gestures in Humorous Discourse by Barack Obama

AU - Navarretta, Costanza

N1 - Conference code: 8

PY - 2017

Y1 - 2017

N2 - In this paper, we aim to predict audience responsefrom simple spoken sequences, speech pauses and co-speechgestures in annotated video- and audio-recorded speeches byBarack Obama at the Annual White House Correspondents’Association Dinner in 2011 and 2016. At these dinners, theAmerican president mocks himself, his collaborators, politicaladversary and the press corps making the audience react withcheers, laughter and/or applause. The results of the prediction experimentdemonstrate that information about spoken sequences,pauses and co-speech gestures by Obama can be used to predictthe immediate audience response. This confirms and shows anapplication of numerous studies that address the importance ofspeech pauses and gestures in delivering the discourse messagein a successful way. The fact that machine learning algorithmscan use information about pauses and gestures to build modelsof audience reaction is also relevant for the construction ofintelligent and cognitively based multimodal ICT.

AB - In this paper, we aim to predict audience responsefrom simple spoken sequences, speech pauses and co-speechgestures in annotated video- and audio-recorded speeches byBarack Obama at the Annual White House Correspondents’Association Dinner in 2011 and 2016. At these dinners, theAmerican president mocks himself, his collaborators, politicaladversary and the press corps making the audience react withcheers, laughter and/or applause. The results of the prediction experimentdemonstrate that information about spoken sequences,pauses and co-speech gestures by Obama can be used to predictthe immediate audience response. This confirms and shows anapplication of numerous studies that address the importance ofspeech pauses and gestures in delivering the discourse messagein a successful way. The fact that machine learning algorithmscan use information about pauses and gestures to build modelsof audience reaction is also relevant for the construction ofintelligent and cognitively based multimodal ICT.

M3 - Article in proceedings

SN - ISBN 978-1-5386-1264-4

SP - 327

EP - 331

BT - 8th IEEE International Conference on Cognitive Infocommunications CogInfoCom 2017 Proceedings

PB - IEEE

T2 - International Conference on Cognitive Infocommunications

Y2 - 11 September 2017 through 14 September 2017

ER -