Speaker Recognition Using Deep Learning Techniques

Accurately classifying age and gender using sound is beneficial for speaker and speech recognition as well as speech emotion classification. This is because separate acoustic models for men and women have been shown to perform better. Gender-based labelling is widely used in image-related applications such as video summary and robot-human interaction. Even still, gender classification is still in its infancy in voice processing and advanced in image processing. Accurate age classification helps determine how best to respond to a given user who fits into a certain group. This can assist customer service organizations and emergency hotlines in tailoring their answer to the individual requiring their help. After meticulous manual selection and using established methods, characteristics retrieved for classification are chosen using machine learning models. To categorize data into the appropriate age and gender classes, we have employed a variety of straightforward classifiers. Additionally, we used various convolutional neural networks with the power spectrum characteristics of the acquired voice samples.