TY - GEN
T1 - Automatic measurement of affective valence and arousal in speech
AU - Asgari, Meysam
AU - Kiss, Geza
AU - Van Santen, Jan
AU - Shafran, Izhak
AU - Song, Xubo
PY - 2014
Y1 - 2014
N2 - Methods are proposed for measuring affective valence and arousal in speech. The methods apply support vector regression to prosodic and text features to predict human valence and arousal ratings of three stimulus types: speech, delexicalized speech, and text transcripts. Text features are extracted from transcripts via a lookup table listing per-word valence and arousal values and computing per-utterance statistics from the per-word values. Prediction of arousal ratings of delexicalized speech and of speech from prosodic features was successful, with accuracy levels not far from limits set by the reliability of the human ratings. Prediction of valence for these stimulus types as well as prediction of both dimensions for text stimuli proved more difficult, even though the corresponding human ratings were as reliable. Text based features did add, however, to the accuracy of prediction of valence for speech stimuli. We conclude that arousal of speech can be measured reliably, but not valence, and that improving the latter requires better lexical features.
AB - Methods are proposed for measuring affective valence and arousal in speech. The methods apply support vector regression to prosodic and text features to predict human valence and arousal ratings of three stimulus types: speech, delexicalized speech, and text transcripts. Text features are extracted from transcripts via a lookup table listing per-word valence and arousal values and computing per-utterance statistics from the per-word values. Prediction of arousal ratings of delexicalized speech and of speech from prosodic features was successful, with accuracy levels not far from limits set by the reliability of the human ratings. Prediction of valence for these stimulus types as well as prediction of both dimensions for text stimuli proved more difficult, even though the corresponding human ratings were as reliable. Text based features did add, however, to the accuracy of prediction of valence for speech stimuli. We conclude that arousal of speech can be measured reliably, but not valence, and that improving the latter requires better lexical features.
KW - affect
KW - arousal
KW - valence
UR - http://www.scopus.com/inward/record.url?scp=84905254292&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84905254292&partnerID=8YFLogxK
U2 - 10.1109/ICASSP.2014.6853740
DO - 10.1109/ICASSP.2014.6853740
M3 - Conference contribution
AN - SCOPUS:84905254292
SN - 9781479928927
T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
SP - 965
EP - 969
BT - 2014 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2014
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2014 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2014
Y2 - 4 May 2014 through 9 May 2014
ER -