RANSAC-Based Training Data Selection on Spectral Features for Emotion Recognition from Spontaneous Speech

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 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 Mel Frequency Cepstral Coefficients (MFCC) and Line Spectral Frequency (LSF) features indicate that utilization of RANSAC in the training phase provides an improvement in the unweighted recall rates on the test set. Experimental studies performed over the FAU Aibo Emotion Corpus demonstrate that decision fusion configurations with LSF and MFCC based classifiers provide further significant performance improvements.

[1]  Björn W. Schuller,et al.  The INTERSPEECH 2009 emotion challenge , 2009, INTERSPEECH.

[2]  Oh-Wook Kwon,et al.  EMOTION RECOGNITION BY SPEECH SIGNAL , 2003 .

[3]  Min Xu,et al.  Efficient sampling of training set in large and noisy multimedia data , 2007, TOMCCAP.

[4]  Eduardo Gasca,et al.  Decontamination of Training Samples for Supervised Pattern Recognition Methods , 2000, SSPR/SPR.

[5]  Isabelle Guyon,et al.  Discovering Informative Patterns and Data Cleaning , 1996, Advances in Knowledge Discovery and Data Mining.

[6]  Thomas G. Dietterich.  Approximate Statistical Tests for Comparing Supervised Classification Learning Algorithms , 1998, Neural Computation.

[7]  Gunnar Rätsch,et al.  Regularizing AdaBoost , 1998, NIPS.

[8]  Pietro Perona,et al.  Pruning training sets for learning of object categories , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[9]  Frank A. Andrews,et al.  Measurements of radiation impedance , 1975 .

[10]  F. Itakura.  Line spectrum representation of linear predictor coefficients of speech signals , 1975 .

[11]  Zhigang Deng,et al.  Emotion recognition based on phoneme classes , 2004, INTERSPEECH.

[12]  Björn W. Schuller,et al.  Patterns, prototypes, performance: classifying emotional user states , 2008, INTERSPEECH.

[13]  Leo Breiman,et al.  Bagging Predictors , 1996, Machine Learning.

[14]  Ibon Saratxaga,et al.  Detection of synthetic speech for the problem of imposture , 2011, 2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[15]  Milan Sonka,et al.  Image Processing, Analysis and Machine Vision , 1993, Springer US.

[16]  Robert C. Bolles,et al.  Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography , 1981, CACM.

[17]  R. Casey,et al.  Advances in Pattern Recognition , 1971 .

[18]  Björn W. Schuller,et al.  Hidden Markov model-based speech emotion recognition , 2003, 2003 International Conference on Multimedia and Expo. ICME '03. Proceedings (Cat. No.03TH8698).

[19]  Leon G. Higley,et al.  Forensic Entomology: An Introduction , 2009 .

[20]  R.W. Morris,et al.  Modification of formants in the line spectrum domain , 2002, IEEE Signal Processing Letters.

[21]  A. Murat Tekalp,et al.  Multimodal speaker identification using an adaptive classifier cascade based on modality reliability , 2005, IEEE Transactions on Multimedia.

[22]  Shrikanth S. Narayanan,et al.  Toward detecting emotions in spoken dialogs , 2005, IEEE Transactions on Speech and Audio Processing.

[23]  Ludmila I. Kuncheva,et al.  Combining Pattern Classifiers: Methods and Algorithms , 2004 .