Feature Selection for Speech Emotion Recognition in Spanish and Basque: On the Use of Machine Learning to Improve Human-Computer Interaction

Abstract

Study of emotions in human–computer interaction is a growing research area. This paper shows an attempt to select the most significant features for emotion recognition in spoken Basque and Spanish Languages using different methods for feature selection. RekEmozio database was used as the experimental data set. Several Machine Learning paradigms were used for the emotion classification task. Experiments were executed in three phases, using different sets of features as classification variables in each phase. Moreover, feature subset selection was applied at each phase in order to seek for the most relevant feature subset. The three phases approach was selected to check the validity of the proposed approach. Achieved results show that an instance-based learning algorithm using feature subset selection techniques based on evolutionary algorithms is the best Machine Learning paradigm in automatic emotion recognition, with all different feature sets, obtaining a mean of 80,05% emotion recognition rate in Basque and a 74,82% in Spanish. In order to check the goodness of the proposed process, a greedy searching approach (FSS-Forward) has been applied and a comparison between them is provided. Based on achieved results, a set of most relevant non-speaker dependent features is proposed for both languages and new perspectives are suggested.

Emotional Speech Recognition Based on the Committee of Classifiers

“…However, a vector of too many features may give rise to high dimension and redundancy, making the learning process complicated and increasing the likelihood of overfitting [ 28 ]. Therefore prior to classification, methods of balancing a numerous features vector, feature selection or extraction are studied to speed up the learning process and minimize the curse of dimensionality problem [ 29 , 30 ]. Emotion classification is generally performed using standard techniques such as SVM [ 31 , 32 , 33 ], various types of artificial neural networks (NN) [ 34 , 35 , 36 , 37 ], different types of the k-NN classifier [ 19 , 38 ] or using Hidden Markov Model (HMM) and its variations [ 39 ].…”

Classifier Subset Selection for the Stacked Generalization Method Applied to Emotion Recognition in Speech

“…Pfister and Robinson [ 30 ] proposed an emotion classification framework that consists of n(n-1)/2 pairwise SVMs for n labels, each with a differing set of features selected by the correlation-based feature selection algorithm. Arruti et al [ 32 ] used four machine learning paradigms (IB, ID3, C4.5, NB) and evolutionary algorithms to select feature subsets that noticeably optimize the automatic emotion recognition success rate. Schuller et al [ 24 ] combined SVMs, decision trees and Bayesian classifiers to yield higher classification accuracy.…”

“…In Table 1 , audio recognition accuracy percentages for the different types of utterances (depending on the language) are presented. It has also to be noted that several automatic emotion recognition systems have used the RekEmozio dataset in previous works, such as [ 32 , 42 ].…”

“…With regard to global features, statistics containing measures, such as the mean, variance, standard deviation and the maximum and minimum values and their positions, were computed, among others. The full set of the 123 features we used in the first phase of this work, including local characteristics and their correlated global statistics, were described in more detail in [ 32 ].…”

Ensemble Learning of Hybrid Acoustic Features for Speech Emotion Recognition

“…There is an avalanche of intrinsic socioeconomic advantages that make speech signals a good source for affective computing. They are economically easier to acquire than other biological signals like electroencephalogram, electrooculography and electrocardiograms [17], which makes speech emotion recognition research attractive [18]. Machine learning algorithms extract a set of speech features with a variety of transformations to appositely classify emotions into different classes.…”