English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Affective computing is very important in the relationship between man andmachine. In this paper, a system for speech emotion recognition (SER) based onspeech signal is proposed, which uses new techniques in different stages ofprocessing. The system consists of three stages: feature extraction, featureselection, and finally feature classification. In the first stage, a complexset of long-term statistics features is extracted from both the speech signaland the glottal-waveform signal using a combination of new and diverse featuressuch as prosodic, spectral, and spectro-temporal features. One of thechallenges of the SER systems is to distinguish correlated emotions. Thesefeatures are good discriminators for speech emotions and increase the SER'sability to recognize similar and different emotions. This feature vector with alarge number of dimensions naturally has redundancy. In the second stage, usingclassical feature selection techniques as well as a new quantum-inspiredtechnique to reduce the feature vector dimensionality, the number of featurevector dimensions is reduced. In the third stage, the optimized feature vectoris classified by a weighted deep sparse extreme learning machine (ELM)classifier. The classifier performs classification in three steps: sparserandom feature learning, orthogonal random projection using the singular valuedecomposition (SVD) technique, and discriminative classification in the laststep using the generalized Tikhonov regularization technique. Also, manyexisting emotional datasets suffer from the problem of data imbalanceddistribution, which in turn increases the classification error and decreasessystem performance. In this paper, a new weighting method has also beenproposed to deal with class imbalance, which is more efficient than existingweighting methods. The proposed method is evaluated on three standard emotionaldatabases.

Speech Emotion Recognition Using Deep Sparse Auto-Encoder Extreme  Learning Machine with a New Weighting Scheme and Spectro-Temporal Features  Along with Classical Feature Selection and A New Quantum-Inspired Dimension  Reduction Method