Publication

Abstract : This work gives a thorough investigation into voice emotion recognition, with a focus on the comparison of machine learning models for this task. It explores the classification of speech data into various emotions, regardless of semantic content, including emotions like happiness, sadness, anger, and neutrality. The primary objective of SER systems is to efficiently detect emotions and incorporate human elements, such as emotional response, into machines. The study investigates and compares several popular machine learning algorithms, including support vector machines (SVM), Classification And Regression Tree (CARTS), Long short-term memory (LSTM). By enabling machines or robots to recognize and respond to emotions, SER systems reduce human time and effort while creating a more empathetic and natural interaction. Overall, this research contributes to the advancement of speech emotion recognition techniques and provides a comparative analysis of machine learning models, enabling researchers and practitioners to make informed decisions when designing and implementing emotion-aware systems.