Hybrid BBO_PSO and higher order spectral features for emotion and stress recognition from natural speech

Yogesh, C K; Hariharan, M.; Ngadiran, Ruzelita; Adom, Abdul Hamid; Yaacob, Sazali; Polat, Kemal

doi:10.1016/j.asoc.2017.03.013

Cited by 41 publications

(8 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The stress recognition model presented in [11] has two stage feature subset selection procedure in which a multi-cluster feature selection is employed in the first stage to identify the relevant features and to minimize the feature space. Biogeography and particle swarm optimization algorithms are incorporated in the second stage to reduce the feature dimensions and effectively differentiate different emotions.…”

Section: Machine Learning Based Emotion Classification Modelsmentioning

confidence: 99%

Stress Level Based Emotion Classification Using Hybrid Deep Learning Algorithm

2023

KSII TIIS

View full text Add to dashboard Cite

The present fast-moving era brings a serious stress issue that affects elders and youngsters. Everyone has undergone stress factors at least once in their lifetime. Stress is more among youngsters as they are new to the working environment. whereas the stress factors for elders affect the individual and overall performance in an organization. Electroencephalogram (EEG) based stress level classification is one of the widely used methodologies for stress detection. However, the signal processing methods evolved so far have limitations as most of the stress classification models compute the stress level in a predefined environment to detect individual stress factors. Specifically, machine learning based stress classification models requires additional algorithm for feature extraction which increases the computation cost. Also due to the limited feature learning characteristics of machine learning algorithms, the classification performance reduces and inaccurate sometimes. It is evident from numerous research works that deep learning models outperforms machine learning techniques. Thus, to classify all the emotions based on stress level in this research work a hybrid deep learning algorithm is presented. Compared to conventional deep learning models, hybrid models outperforms in feature handing. Better feature extraction and selection can be made through deep learning models. Adding machine learning classifiers in deep learning architecture will enhance the classification performances. Thus, a hybrid convolutional neural network model was presented which extracts the features using CNN and classifies them through machine learning support vector machine. Simulation analysis of benchmark datasets demonstrates the proposed model performances. Finally, existing methods are comparatively analyzed to demonstrate the better performance of the proposed model as a result of the proposed hybrid combination.

show abstract

Section: Machine Learning Based Emotion Classification Modelsmentioning

confidence: 99%

Stress Level Based Emotion Classification Using Hybrid Deep Learning Algorithm

2023

KSII TIIS

View full text Add to dashboard Cite

show abstract

“…Instead of treating each speaker and emotion separately, the model jointly optimize feature selection across multiple tasks, and it identifies the features that are relevant for emotion recognition of different speakers. Yogesh et al used high order spectral analysis (HOSA) to extract BSF and BCF from speech [22]. By combining these features with the standard 2010 interspeech features, the performance of a real-time SER system is further enhanced.…”

Section: Related Workmentioning

confidence: 99%

“…To validate the superiority of the proposed BiEO, the classification performance is compared with EO [28], BBO_PSO [22] and MOBFA [30]. EO is utilized to test whether the proposed BiEO enhances the performance of classical EO.…”

Section: B Experimental Setupmentioning

confidence: 99%

Equilibrium Optimizer for Emotion Classification From English Speech Signals

Yue,

Hu,

Chu

et al. 2023

IEEE Access

View full text Add to dashboard Cite

Speech emotion recognition and its precise classification are challenging tasks that heavily depend on the quality of feature extraction and selection for speech signals. Many feature selection algorithms have been proposed to achieve recognition, however, their accuracy has not reached a satisfactory level. We introduce an improved equilibrium optimizer (iEO) algorithm and utilize mel frequency cepstral coefficients (MFCCs) and pitch features for emotion recognition. The transfer function is used to complete the binarization of iEO (BiEO), and the algorithm adopts multi-swarm and transfer functions to balance global search and local search. The performance of the proposed algorithm is verified using four English speech emotion datasets, eNTERFACE05, ryerson audio-visual database of emotional speech and song (RAVDESS), surrey audio-visual expressed emotion (SAVEE) and toronto emotional speech set (TESS). The experimental results illustrate that the proposed algorithm obtains an accuracy of 0.4923, 0.5581, 0,5575 and 0.9840 in eNTERFACE05, RAVDESS, SAVEE and TESS based on K-nearest neighbors, and an accuracy of 0.5279, 0.5862, 0.6752 and 0.9941 based on random forest.

show abstract

“…Later, specifically for anger emotion recognition, acoustic (pitch, loudness, spectral features) and linguistic (probabilistic and entropy-based words and phrases) cues [ 6 ] were introduced. Apart from these, other different feature extraction techniques such as a sinusoidal model-based feature extraction technique with frequency, magnitude, and phase features [ 7 ]; empirical mode decomposition method; feature optimization method [ 8 ] to select particular frames of the speech signal by choosing proper filter bank; hybrid biogeography-based optimization; and particle swarm optimization (BBO_PSO) [ 9 ] by the proper selection of higher-order spectral features were used for depressed emotion recognition. Yang and Lugger [ 10 ] proposed the combination of qualitative and voice quality features; Wen et al [ 11 ] proposed weighted spectral local Hu parameters to overcome the disadvantage of MFCC feature; Wang et al [ 12 ] proposed Fourier parameters; Setayeshi et al [ 13 ] proposed a bioinspired ANFIS technique combined with MLP for SER for anger, happy and sad emotions; and, Ying and Xue-Ying [ 14 ] proposed glottal compensation to zero crossings with maximal Teager energy operator (GCZCMT) for speech emotion recognition and performed well compared to MFCC feature.…”

Section: Introductionmentioning

confidence: 99%

Stressed Speech Emotion Recognition Using Teager Energy and Spectral Feature Fusion with Feature Optimization

Bandela,

Siva Priyanka,

Sunil Kumar

et al. 2023

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

The objective of speech emotion recognition (SER) is to enhance man–machine interface. It can also be used to cover the physiological state of a person in critical situations. In recent time, speech emotion recognition also finds its operations in medicine and forensics. A new feature extraction technique using Teager energy operator (TEO) is proposed for the detection of stressed emotions as Teager energy-autocorrelation envelope (TEO-Auto-Env). TEO is basically designed for increasing the energies of the stressed speech signals whose energies are reduced during the speech production process and hence used in this analysis. A stressed speech emotion recognition (SSER) system is developed using TEO-Auto-Env and spectral feature combination for detecting the emotions. The spectral features considered are Mel-frequency cepstral coefficients (MFCC), linear prediction cepstral coefficients (LPCC), and relative spectra–perceptual linear prediction (RASTA-PLP). EMO-DB (German), EMOVO (Italian), IITKGP (Telugu), and EMA (English) databases are used in this analysis. The classification of the emotions is carried out using the k-nearest neighborhood (k-NN) classifier for gender-dependent (GD) and speaker-independent (SI) cases. The proposed SSER system provides improved accuracy compared to the existing ones. Average recall is used for performance evaluation. The highest classification accuracy is achieved using the feature combination of TEO-Auto-Env, MFCC, and LPCC features with 91.4% (SI), 91.4% (GD-male), and 93.1%(GD-female) for EMO-DB; 68.5% (SI), 68.5% (GD-male), and 74.6% (GD-female) for EMOVO; 90.6%(SI), 91% (GD-male), and 92.3% (GD-female) for EMA; and 95.1% (GD-female) for IITKGP female database.

show abstract

Hybrid BBO_PSO and higher order spectral features for emotion and stress recognition from natural speech

Cited by 41 publications

References 49 publications

Stress Level Based Emotion Classification Using Hybrid Deep Learning Algorithm

Stress Level Based Emotion Classification Using Hybrid Deep Learning Algorithm

Equilibrium Optimizer for Emotion Classification From English Speech Signals

Stressed Speech Emotion Recognition Using Teager Energy and Spectral Feature Fusion with Feature Optimization

Contact Info

Product

Resources

About