Speaker-independent negative emotion recognition

Abstract

Abstract-This work aims to provide a method able to distinguish between negative and non-negative emotions in vocal interaction. A large pool of 1418 features is extracted for that purpose. Several of those features are tested in emotion recognition for the first time. Next, feature selection is applied separately to male and female utterances. In particular, a bidirectional Best First search with backtracking is applied. The first contribution is the demonstration that a significant number of features, first tested here, are retained after feature selection. The selected features are then fed as input to support vector machines with various kernel functions as well as to the K nearest neighbors classifier. The second contribution is in the speaker-independent experiments conducted in order to cope with the limited number of speakers present in the commonly used emotion speech corpora. Speakerindependent systems are known to be more robust and present a better generalization ability than the speaker-dependent ones. Experimental results are reported for the Berlin emotional speech database. The best performing classifier is found to be the support vector machine with the Gaussian radial basis function kernel. Correctly classified utterances are 86.73%±3.95% for male subjects and 91.73%±4.18% for female subjects. The last contribution is in the statistical analysis of the performance of the support vector machine classifier against the K nearest neighbors classifier as well as the statistical analysis of the various support vector machine kernels impact.

Speaker-independent emotion recognition exploiting a psychologically-inspired binary cascade classification schema

“…The latter is just the first level of the proposed psychologicallyinspired binary cascade classification schema presented in Figure 1. The best accuracy achieved in [32] equals about 90.0%. Here with the extraction of 2327 features the corresponding accuracy (i.e.…”

“…With respect to our previous work [32], a smaller set of 1418 features was computed to discriminate between negative and non-negative valence emotions. The latter is just the first level of the proposed psychologicallyinspired binary cascade classification schema presented in Figure 1.…”

“…Mean of the 1st, 2nd, 3rd, and 4th formant [5][6][7][8] Maximum of the 1st, 2nd, 3rd, and 4th formant [9][10][11][12] Minimum of the 1st, 2nd, 3rd, and 4th formant [13][14][15][16] Variance of the 1st, 2nd, 3rd, and 4th formant [17][18][19][20] Skewness of the 1st, 2nd, 3rd, and 4th formant [21][22][23][24] Interquartile range of the 1st, 2nd, 3rd, and 4th formant [25][26][27][28] Range (i.e. maximum-minimum) of the 1st, 2nd, 3rd, and 4th formant [29][30][31][32] 90th percentile of the 1st, 2nd, 3rd, and 4th formant Several observations are made by the research community with respect to the correspondence among features and emotional categories. Previous work [14] has indicated that anger presents an increased pitch, happiness demonstrates an increased intensity, whereas sadness exhibits a decrease in high-frequency energy.…”

Recognition of Human Emotion from a Speech Signal Based on Plutchik's Model

“…In terms of range of emotions, most researchers concern the detection of negative and positive emotions, without indicating a specific state [5] or focus on aggression detection [6]. There are also many studies investigating the so-called basic emotions like happiness, anger, suprise, sadness, fear, disgust and neutral state, which are most commonly used because of their availability in public databases.…”

Análise de sinais de voz para reconhecimento de emoções.

“…Apesar de ser uma base de dados composta por emoções atuadas, neste momento optamos por uma base estável e popular, que já foi utilizada em diversos estudos (37)(38)(64) (6)(68)(45) (65). Uma vez que a comparação entre resultados na literatura é difícil já que existem muitas variáveis envolvidas, a utilização de uma base conhecida e bastante estudada facilita a comparação de resultados, pois elimina algumas destas variáveis, como as emoções escolhidas, o número de emoções, além das próprias amostras de áudio.…”