Speech Emotion Recognition Using Regularized Discriminant Analysis

Kuchibhotla, Swarna; Yalamanchili, Bhanusree; Vankayalapati, Hima Deepthi; Anne, Koteswara Rao

doi:10.1007/978-3-319-02931-3_41

Cited by 8 publications

(4 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…6 Bad <0. 5 Test not useful Table2 in [10] are satisfactory if it is less than 0.5 the classification technique is not useful.…”

Section: Resultsmentioning

confidence: 97%

“…LDA is a statistical method which "reduces the dimension of the features while maximizing the information preserved in the reduced feature space" [4]. LDA minimizes the within class covariance and maximizes the between class covariance in the given data base of speech samples thereby guaranteeing maximal separability [11] [6]. The general procedure for implementing the LDA is shown in But in this paper we used two different approaches for implementing LDA, Class Dependent LDA and Class Independent LDA.…”

Section: E Discrete Cosine Transformmentioning

confidence: 99%

See 1 more Smart Citation

ROC Analysis of Class Dependent and Class Independent Linear Discriminant classifiers using frequency domain features

Kuchibhotla

Vankayalapati²,

Yalamanchili³

et al. 2014

2014 International Conference on Advances in Computing, Communications and Informatics (ICACCI)

Self Cite

View full text Add to dashboard Cite

Emotional speech recognition aims at classifying the human emotional states viz. happy, neutral, anger and sad etc.,.To classify these emotions we need to extract reliable Acoustic features like prosody and spectral. The time domain features are much less accurate than frequency domain features . So in this paper Mel Frequency Cepstral Coefficients(MFCC) are extracted from Berlin emotional speech corpus and are classified using Class Dependent and Class Independent Linear Discriminant Analysis(CD-LDA and CI-LDA). The results obtained shows the performance variation of the classifiers with respect to the emotional states.

show abstract

“…6 Bad <0. 5 Test not useful Table2 in [10] are satisfactory if it is less than 0.5 the classification technique is not useful.…”

Section: Resultsmentioning

confidence: 97%

Section: E Discrete Cosine Transformmentioning

confidence: 99%

ROC Analysis of Class Dependent and Class Independent Linear Discriminant classifiers using frequency domain features

Kuchibhotla

Vankayalapati²,

Yalamanchili³

et al. 2014

2014 International Conference on Advances in Computing, Communications and Informatics (ICACCI)

Self Cite

View full text Add to dashboard Cite

show abstract

“…K-Nearest Neighbor is a non-parametric method that classifies speech dataset based on closest training samples in the feature space [2]. In the classification phase, an unlabeled vector is classified by assigning the label which is most frequent among the k training samples nearest to the k point where k is a user-defined constant, Unclassified speech samples are sent into the system to extract speech coefficients and uses model file to classify the speech emotion [19].…”

Section: K-nearest Neighbor (Knn)mentioning

confidence: 99%

Speech Emotion Recognition: A Survey

Kuchibhotla¹

2019

IJMUE

View full text Add to dashboard Cite

Speech Recognition is a tremendous application from the history that is identification and conversion of spoken words into text. The performance and quality of work had increased a lot. This performance lead to the research work on Emotion Recognition based on the language spoken that is obtaining the kind of emotion from the spoken speech which is an application based on human-robot interactions. Emotions can be recognized in a better way using Speech processing, Artificial Intelligence techniques and linguistic semantics. Systems are given training in such a way to detect the emotions from the spoken utterances. This paper contains about the survey from the history to the present works that took place in the speech emotion recognition and also the experiment results. The survey contains about the works that took place from the by different scientists and their usage of different features, classifiers etc. The paper also holds three categories, one is different databases that are involved, second is what features are involved for representation of speech and third about the classification schemes. The survey also includes the conclusions of performances and limitations of current speech emotion recognition.

show abstract

“…The first stage of feature extraction and selection can be performed by handpicking features or using neural networks. Previous studies have used prosodic, spectral data, Mel Frequency Cepstral Coefficient (MFCC), Linear prediction cepstral coefficients (LPCC) and formant features (Gudmalwar et al, 2019;Kuchibhotla et al, 2014;Lalitha et al, 2019;Pawar & Kokate, 2021). Training neural networks to extract appropriate features rather than handpicking will solve most of the issues related to spatial-temporal features .…”

Section: Introductionmentioning

confidence: 99%

Time-Distributed Attention-Layered Convolution Neural Network with Ensemble Learning using Random Forest Classifier for Speech Emotion Recognition

Bhanusree

Kumar

Rao

2023

Journal of Information and Communication Technology

View full text Add to dashboard Cite

Speech Emotion Detection (SER) is a field of identifying human emotions from human speech utterances. Human speech utterancesare a combination of linguistic and non-linguistic information. Nonlinguistic SER provides a generalized solution in human–computerinteraction applications as it overcomes the language barrier. Machine learning and deep learning techniques were previously proposed for classifying emotions using handpicked features. To achieve effective and generalized SER, feature extraction can be performed using deep neural networks and ensemble learning for classification. The proposed model employed a time-distributed attention-layered convolution neural network (TDACNN) for extracting spatiotemporal features at the first stage and a random forest (RF) classifier, which is an ensemble classifier for efficient and generalized classification of emotions, at the second stage. The proposed model was implemented on the RAVDESS and IEMOCAP data corpora and compared with the CNN-SVM and CNN-RF models for SER. The TDACNN-RF model exhibited test classification accuracies of 92.19 percent and 90.27 percent on the RAVDESS and IEMOCAP data corpora, respectively. The experimental results proved that the proposed model is efficient in extracting spatiotemporal features from time-series speech signals and can classify emotions with good accuracy. The class confusion among the emotions was reduced for both data corpora, proving that the model achieved generalization.

show abstract

Speech Emotion Recognition Using Regularized Discriminant Analysis

Cited by 8 publications

References 7 publications

ROC Analysis of Class Dependent and Class Independent Linear Discriminant classifiers using frequency domain features

ROC Analysis of Class Dependent and Class Independent Linear Discriminant classifiers using frequency domain features

Speech Emotion Recognition: A Survey

Time-Distributed Attention-Layered Convolution Neural Network with Ensemble Learning using Random Forest Classifier for Speech Emotion Recognition

Contact Info

Product

Resources

About