research-article

RANSAC-based training data selection for emotion recognition from spontaneous speech

Authors:

Cigdem Eroglu Erdem,

A. Tanju ErdemAuthors Info & Claims

AFFINE '10: Proceedings of the 3rd international workshop on Affective interaction in natural environments

Pages 9 - 14

https://doi.org/10.1145/1877826.1877831

Published: 29 October 2010 Publication History

Abstract

Training datasets containing spontaneous emotional expressions are often imperfect due the ambiguities and difficulties of labeling such data by human observers. In this paper, we present a Random Sampling Consensus (RANSAC) based training approach for the problem of emotion recognition from spontaneous speech recordings. Our motivation is to insert a data cleaning process to the training phase of the Hidden Markov Models (HMMs) for the purpose of removing some suspicious instances of labels that may exist in the training dataset. Our experiments using HMMs with various number of states and Gaussian mixtures per state indicate that utilization of RANSAC in the training phase provides an improvement of up to 2.84% in the unweighted recall rates on the test set. . This improvement in the accuracy of the classifier is shown to be statistically significant using McNemar's test.

References

[1]

Angelova, A., Abu-Mostafa, Y., and Perona, P. 2005. Pruning Training Sets for Learning of Object Categories, Proc. Int. Conf. on Computer Vision and Pattern Recognition (CVPR).

Digital Library

[2]

Barandela, R., and Gasca, E. 2000, Decontamination of Training Samples for Supervised Pattern Recognition Methods. Lecture Notes in Computer Science, vol. 1876, pp. 621--630.

Digital Library

[3]

Ben-Gal I., Outlier detection, In: Maimon O. and Rockach L. (Eds.) Data Mining and Knowledge Discovery Handbook: A Complete Guide for Practitioners and Researchers, Kluwer Academic Publishers, 2005.

[4]

Bozkurt, E., Erzin, E., Erdem, C. E., Erdem, A. T. 2009. Improving Automatic Emotion Recognition from Speech Signals. Interspeech 2009, ISCA.

[5]

Bozkurt, E., Erdem, C. E., Erdem, A. T., Erzin, E. 2010. Use of Line Spectral Frequencies for Emotion Recognition from Speech. Int. Conf. on Pattern Recognition, August 2010, 0stanbul, Turkey.

Digital Library

[6]

Breiman, L., 1996. Bagging Predictors. Machine Learning, 24(2), 123--140.

Digital Library

[7]

Dietterich, T. G., Approximate Statistical Tests for Comparing Supervised Classification Learning Algorithms. Neural Computation, 7(10):1895--1924, 1998.

Digital Library

[8]

Fischler, M. A., and Bolles, R. C. 1981. Random Sample Consensus: A Paradigm for Model Fitting with Applications to Image Analysis and Automated Cartography. Graphics and Image Processing, Vol. 24, No. 6.

Digital Library

[9]

Gu, B., Hu, F., Liu, H. 2000. Sampling and its Applications in Data Mining: A Survey. Tech. Rep. School of Computing, National University of Singapore, Singapore.

[10]

Guyon, I., Matin, N., Vapnik, V. 1994. Discovering informative Patterns and Data Cleaning. Workshop on Knowledge Discovery in Databases.

[11]

Kuncheva, L. I. Combining Pattern Classifiers. John Wiley & Sons, 2004.

Digital Library

[12]

Olken, F. 1993. Random Sampling From Databases. Ph. D. Thesis, Department of Computer Science, University of California, Berkeley.

[13]

Ratsch, G., Onoda, T., and Muller, K. 2000. Regularizing Adaboost, Advances in Neural Information Processing Systems, vol. 11, 564--570.

Digital Library

[14]

Schuller, B., Steidl, S., and Batliner, A. 2009. The INTERSPEECH 2009 Emotion Challenge. Interspeech (2009), ISCA.

[15]

Seppi, D., Batliner, A., Schuller, B., Steidl, S., Vogt, T., Wagner, J., Devillers, L., Vidrascu, L., Amir, N., Aharonson, V.2008. Patterns, Prototypes, Performance: Classifying Emotional User States. Interspeech (2008), ISCA.

[16]

Sonka, M., Hlavac, V., and Boyle, R. 2008. Image Processing, Analysis and Machine Vision, Thomson.

Digital Library

[17]

Wang, S., Dash, M., Chia, L. and Xu, M. 2007. Efficient sampling of training set in large and noisy multimedia data. ACM Transactions on Multimedia Computing, Communications and Applications, Vol. 3, No.3.

Digital Library

[18]

The Hidden Markov Toolkit, http://htk.eng.cam.ac.uk/

Cited By

Antenore MCamacho Rodriguez JPanizzi E(2022)A Comparative Study of Bot Detection Techniques With an Application in Twitter Covid-19 DiscourseSocial Science Computer Review10.1177/0894439321107373341:5(1520-1545)Online publication date: 5-Apr-2022
https://doi.org/10.1177/08944393211073733
Rizos GSchuller B(2019)Modelling Sample Informativeness for Deep Affective ComputingICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)10.1109/ICASSP.2019.8683729(3482-3486)Online publication date: May-2019
https://doi.org/10.1109/ICASSP.2019.8683729
Mustafa MYusoof MDon ZMalekzadeh M(2018)Speech emotion recognition researchInternational Journal of Speech Technology10.1007/s10772-018-9493-x21:1(137-156)Online publication date: 1-Mar-2018
https://dl.acm.org/doi/10.1007/s10772-018-9493-x
Show More Cited By

Index Terms

RANSAC-based training data selection for emotion recognition from spontaneous speech
1. Hardware
  1. Communication hardware, interfaces and storage
    1. Signal processing systems

Recommendations

RANSAC-based training data selection on spectral features for emotion recognition from spontaneous speech
COST'10: Proceedings of the 2010 international conference on Analysis of Verbal and Nonverbal Communication and Enactment

Training datasets containing spontaneous emotional speech are often imperfect due the ambiguities and difficulties of labeling such data by human observers. In this paper, we present a Random Sampling Consensus (RANSAC) based training approach for the ...
Fear emotion classification in speech by acoustic and behavioral cues

Machine-based emotional speech classification has become a requirement for natural and familiar human-computer interactions. Because emotional speech recognition systems use a person's voice to spontaneously detect their emotional status and take ...
Class-level spectral features for emotion recognition

The most common approaches to automatic emotion recognition rely on utterance-level prosodic features. Recent studies have shown that utterance-level statistics of segmental spectral features also contain rich information about expressivity and emotion. ...

Comments

Information & Contributors

Information

Published In

cover image ACM Conferences

AFFINE '10: Proceedings of the 3rd international workshop on Affective interaction in natural environments

October 2010

106 pages

ISBN:9781450301701

DOI:10.1145/1877826

Program Chairs:
Ginevra Castellano
Queen Mary University of London, UK
,
Kostas Karpouzis
National Technical University of Athens, Greece
,
Jean-Claude Martin
LIMSI-CNRS, France
,
Louis-Philippe Morency
University of Southern California, USA
,
Christopher Peters
Coventry University, UK
,
Laurel Riek
University of Cambridge, UK

Copyright © 2010 ACM.

Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected]

Sponsors

SIGMM: ACM Special Interest Group on Multimedia

Publisher

Association for Computing Machinery

New York, NY, United States

Publication History

Published: 29 October 2010

Permissions

Request permissions for this article.

Request Permissions

Check for updates

Author Tags

Qualifiers

Research-article

Conference

MM '10

Sponsor:

SIGMM

MM '10: ACM Multimedia Conference

October 29, 2010

Firenze, Italy

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

10
Total Citations
View Citations
173
Total Downloads

Downloads (Last 12 months)3
Downloads (Last 6 weeks)0

Reflects downloads up to 30 Dec 2024

Other Metrics

View Author Metrics

Citations

Cited By

Antenore MCamacho Rodriguez JPanizzi E(2022)A Comparative Study of Bot Detection Techniques With an Application in Twitter Covid-19 DiscourseSocial Science Computer Review10.1177/0894439321107373341:5(1520-1545)Online publication date: 5-Apr-2022
https://doi.org/10.1177/08944393211073733
Rizos GSchuller B(2019)Modelling Sample Informativeness for Deep Affective ComputingICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)10.1109/ICASSP.2019.8683729(3482-3486)Online publication date: May-2019
https://doi.org/10.1109/ICASSP.2019.8683729
Mustafa MYusoof MDon ZMalekzadeh M(2018)Speech emotion recognition researchInternational Journal of Speech Technology10.1007/s10772-018-9493-x21:1(137-156)Online publication date: 1-Mar-2018
https://dl.acm.org/doi/10.1007/s10772-018-9493-x
Zhang ZCummins NSchuller B(2017)Advanced Data Exploitation in Speech Analysis: An overviewIEEE Signal Processing Magazine10.1109/MSP.2017.269935834:4(107-129)Online publication date: Jul-2017
https://doi.org/10.1109/MSP.2017.2699358
Kim Y(2015)Exploring sources of variation in human behavioral dataProceedings of the 2015 International Conference on Affective Computing and Intelligent Interaction (ACII)10.1109/ACII.2015.7344653(748-753)Online publication date: 21-Sep-2015
https://dl.acm.org/doi/10.1109/ACII.2015.7344653
Le DProvost E(2015)Data selection for acoustic emotion recognitionProceedings of the 2015 International Conference on Affective Computing and Intelligent Interaction (ACII)10.1109/ACII.2015.7344564(146-152)Online publication date: 21-Sep-2015
https://dl.acm.org/doi/10.1109/ACII.2015.7344564
Zhalehpour SAkhtar ZEroglu Erdem C(2015)Multimodal emotion recognition based on peak frame selection from videoSignal, Image and Video Processing10.1007/s11760-015-0822-010:5(827-834)Online publication date: 29-Sep-2015
https://doi.org/10.1007/s11760-015-0822-0
Kim STakahashi STanaka J(2015)A location-sensitive visual interface on the palmPersonal and Ubiquitous Computing10.1007/s00779-014-0769-019:1(175-187)Online publication date: 1-Jan-2015
https://dl.acm.org/doi/10.1007/s00779-014-0769-0
Weninger FWöllmer MSchuller B(2015)Emotion Recognition in Naturalistic Speech and Language—A SurveyEmotion Recognition10.1002/9781118910566.ch10(237-267)Online publication date: 2-Jan-2015
https://doi.org/10.1002/9781118910566.ch10
Schuller BWeninger F(2011)Ten recent trends in computational paralinguisticsProceedings of the 2011 international conference on Cognitive Behavioural Systems10.1007/978-3-642-34584-5_3(35-49)Online publication date: 21-Feb-2011
https://dl.acm.org/doi/10.1007/978-3-642-34584-5_3

View Options

Login options

Check if you have access through your login credentials or your institution to get full access on this article.

Full Access

Get this Publication

View options

PDF

View or Download as a PDF file.

eReader

View online with eReader.

Media

Figures

Other

Tables

View Table of Contents