Characteristics and recognizability of vocal expressions of emotion

176 Pages·1984·8.448 MB·English

by  Bezooijen, Renée van

Characteristics
and Recognizability
of Vocal Expressions
of Emotion
Netherlands Phonetic Archives
The Netherlands Phonetic Archives (NPA) are modestly priced series of
monographs or edited volumes of papers, reporting recent advances in the
field of phonetics and experimental phonology. The archives address an
audience of phoneticians, phonologists and psycholinguists.
Editors
Marcel P.R. Van den Broecke
University of Utrecht
Vincent J. van Heuven
University of Leyden
Other books in this series:
I Nico Willems
English Intonation from a Dutch Point of View
IIA A. Cohen and M.P.R. Van den Broecke (eds.)
Abstracts of the Tenth International Congress of Phonetic Sciences
IIB M.P.R. Van den Broecke and A. Cohen (eds.)
Proceedings of the Tenth International Congress of Phonetic Sciences
III J.R. de Pijper
Modelling British English Intonation
IV Lou Boves
The Phonetic Basis of Perceptual Rating? of Running Speech
Renee van Bezooyen
Characteristics
and Recognizability
of Vocal Expressions
of Emotion
¥
1984
FORIS PUBLICATIONS
Dordrecht - Holland/Cinnaminson - U.SA.
Published by:
Foris Publications Holland
P.O. Box 509
3300 AM Dordrecht, The Netherlands
Sole distributor for the U.SA. and Canada:
Foris Publications U.SA
P.O. Box C-50
Cinnaminson N.J. 08077
U.SA
CIP
Frontcover illustration taken from F.M. Helmont [An unbreviated representation of a true
natural Hebrew alphabeth, which simultaneously shows how those born deaf can be thought
not only to understand others who speak but even to produce speech themselves], Pieter
Rotterdam, Amsterdam 1697.
ISBN 90 6765 085 4(Bound)
ISBN 90 6765 084 6(Paper)
© 1984 Foris Publications - Dordrecht.
No part of this publication may be reproduced or transmitted in any form or by any means,
electronic or mechanical, including photocopy, recording, or any information storage and
retrieval system, without permission from the copyright owner.
Printed in the Netherlands by ICG Printing, Dordrecht.
CONTENTS
1. INTRODUCTION 1
1.1 Introduction 1
1.2 Description 2
1.3 Recognition 3
1.4 Categorical and dimensional approach 4
1.5 Structure of the book 5
2. SPEECH MATERIAL 7
2.1 Introduction 7
2.2 Method 13
2.2.1 Speakers 13
2.2.2 Phrases 13
2.2.3 Instructions 14
2.2.4 Recordings 14
3. PERCEPTUAL CHARACTERISTICS OF DISCRETE EMO-
TIONS 17
3.1 Introduction 17
3.2 Method 19
3.2.1 Speech material 19
3.2.2 Raters 19
3.2.3 Parameters 19
3.2.4 Rating 21
3.3 Results 21
3.3.1 Reliability 22
3.3.2 Factors sex of speaker, phrase, and emotion 22
3.3.2.1 Factor sex of speaker 22
3.3.2.2 Factor phrase 24
3.3.2.3 Factor emotion 24
3.4 Discussion 28
3.4.1 Reliability 28
3.4.2 Factor sex of speaker 29
3.4.3 Factor phrase 30
3.4.4 Perceptual characteristics of emotions 31
VI
4. RELATIVE IMPORTANCE OF PERCEPTUAL PARAME-
TERS AND THE CHARACTERISTICS OF EMOTIONAL DI-
MENSIONS 39
4.1 Introduction 39
4.2 Method 42
4.2.1 Input data 42
4.2.2 Analysis 42
4.3 Results 43
4.3.1 Classification 43
4.3.2 Dimensions 46
4.3.3 Confusions 49
4.4 Discussion SI
4.4.1 Classification 51
4.4.2 Relative importance of perceptual parameters 52
4.4.3 Dimensions 53
5. ACOUSTIC CORRELATES OF PERCEPTUAL PARAMETERS 57
5.1 Introduction 57
5.2 Method 58
5.2.1 Perceptual ratings 58
5.2.2 Measures of tempo 58
5.2.3 The measurement of fundamental frequency 59
5.2.3.1 Measures of central tendency of F 59
0
5.2.3.2 Measures of F„ variation 60
5.2.3.3 Measures of F perturbations 60
0
5.2.4 Measures of spectral slope 61
5.3 Results 64
5.3.1 Ratings of tempo related to syllable duration 64
5.3.2 Ratings of pitch level explained by measures of central
tendency of F 64
0
5.3.3 Ratings of pitch range explained by measures of F varia-
0
tion 65
5.3.4 Ratings of harshness explained by measures of F pertur-
0
bations 66
5.3.5 Ratings of loudness, tension, and laxness explained by
measures of spectral slope 66
5.3.6 A multiple regression approach 67
5.3.7 Discriminant analysis 69
5.4 Discussion 70
6. RECOGNITION BY DUTCH ADULTS 75
6.1 Introduction 75
6.2 Method 83
6.2.1 Subjects 83
6.2.2 Speech material 83
VII
6.2.3 Procedure 83
6.3 Results 84
6.3.1 Factor emotion 84
6.3.2 Factors sex of speaker and sex of subject 86
6.3.3 Factor phrase 86
6.4 Discussion 87
6.4.1 Factor emotion 87
6.4.2 Factors sex of speaker and sex of subject 89
6.4.3 Factor phrase 90
7. DEVELOPMENTAL RECOGNITION 91
7.1 Introduction 91
7.2 Method 96
7.2.1 Subjects 96
7.2.2 Speech material 97
7.2.3 Instruments 98
7.2.4 Response categories 98
7.2.5 Procedure 100
7.3 Results 102
7.3.1 Factorage 102
7.3.2 Factor sex of subject 104
7.3.3 Factor test 104
7.3.4 Confusions 105
7.4 Discussion 106
7.4.1 Factorage 106
7.4.2 Factor sex of subject 108
7.4.3 Factor test 109
7.4.4 Confusions 112
8. RECOGNITION BY TODDLERS 113
8.1 Introduction 113
8.2 Method 114
8.2.1 Subjects 114
8.2.2 Speech material 114
8.2.3 Instruments 115
8.2.4 Response categories 115
8.2.5 Procedure 115
8.3 Results 117
8.4 Discussion 119
9. RECOGNITION BY DUTCH, TAIWANESE, AND JAPANESE
SUBJECTS 121
9.1 Introduction 121
9.2 Method 124
9.2.1 Subjects 124
VIII
9.2.2 Speech material 124
9.2.3 Procedure 124
9.3 Results 125
9.3.1 Rate of recognition 126
9.3.2 Confusions 129
9.4 Discussion 137
lO.INTEGRATION OF FINDINGS AND SUGGESTIONS FOR
FURTHER RESEARCH 141
10.1 Introduction 141
10.2 Cross-cultural similarity of discrete emotions 141
10.3 Emotional dimensions 142
10.4 Rate of recognition 145
10.5 Some methodological remarks and suggestions for further
research 147
APPENDIX A. Acoustic data for ten emotions 149
APPENDIX B. Instructions for the recognition experiment with
Dutch adults 150
APPENDIX C. Combinations and order of the alternatives in the
developmental recognition experiment 152
APPENDIX D. Combinations and order of the alternatives in the
recognition experiment with toddlers 152
REFERENCES 153
S AMEN V ATTING 159
CURRICULUM VITAE 165
Acknowledgements
The research reported on in this study was funded by the Research Pool of the
Catholic University of Nijmegen on the basis of a three year contract; it was
carried out at the Institute of Phonetics of this University.
I am particularly grateful to Loe Boves for his interest, his patience, and his
invariably sensible comments, to Bert Cranen, for his help in carrying out the
acoustic analyses, and to Toni Rietveld, for his statistical advice.
I would like to thank my student-assistents Wim Hoogeveen for the time he
saved me in carrying out those countless little jobs that research involves and
Annie van de Oever for het ever present, stimulating interest and the devotion
with which she handled each activity she undertook.
I would further like to thank Hans Zondag, who made the recordings, Stan
Otto and Tom Heenan, who conducted the cross-cultural experiments, and all
those children and adults who in one way or another participated in this
research, either as subjects, speakers, or raters.
Finally, I wish to express my gratitude to Joe for going so patiently and
carefully over my English.