Publication Date: 2023/08/25
Abstract: Speech recognition has gained significant importance in facilitating user interactions with various technologies. Recognizing human emotions and affective states from speech, known as Speech Emotion Recognition (SER), has emerged as a rapidly growing research subject. Unlike humans, machines lack the innate ability to perceive and express emotions. Therefore, leveraging speech signals for emotion detection has become an adaptable and accessible approach. This paper presents a project aimed at classifying emotional states in speech for applications such as call centers, measuring emotional attachment in phone calls, and real-time emotion recognition in online learning. The classification methods employed in this study include Support Vector Machines (SVM), Logistic Regression (LR), and Multi-Layer Perceptron (MLP). The project utilizes features such as Mel-frequency cepstrum coefficients (MFCC), chroma, and mel to extract relevant information from speech signals and train the classifiers. Through a comparative analysis of these classification methods, this research aims to enhance the understanding of speech emotion recognition and contribute to the development of more effective and accurate emotion recognition systems.
Keywords: Speech Emotion Recognition, Speech Recognition (SER), Emotion Classification, Support Vector Machines (SVM), Logistic Regression (LR), Multi-Layer Perceptron (MLP), Mel-frequency Cepstrum Coefficients (MFCC), Chroma, Mel Features.
DOI: https://doi.org/10.5281/zenodo.8282877
PDF: https://ijirst.demo4.arinfotech.co/assets/upload/files/IJISRT23MAY770.pdf
REFERENCES