Prediction of Gender and Emotion Using Acoustic Features
S. Sathyavathi, H. Deksha, Ajay Krishnan T, M. Santhosh · 2023
Detecting gender and emotion from voice data is a challenging issue in machine learning. This paper presents a novel approach for detecting gender and emotion from voice data using machine learning techniques. The proposed system uses a dataset of voice recordings that have been preprocessed and labeled according to their gender and emotion categories. Using Mel Frequency Cepstral Coefficients, the feature extraction phase is carried out. To build a classification model that can accurately predict the gender and emotion of a new voice sample, different machine learning algorithms are used. The predicted algorithm is evaluated on a large database of voice recordings, and the analysis shows that it can predict gender and emotion with improved accuracy using Convolutional Neural Network with Long Short-Term Memory. The proposed approach has several potential applications in areas such as speech recognition, voice-based authentication.