Deep Learning Techniques for Voice-Based Gender and Age Prediction

This paper delves into deep learning approaches for gender and age prediction from voice data. We discuss a comparison of logistic regression and long short-term memory networks to predict gender, with Kmeans clustering along with logistic regression to predict the age of the speaker. In the obtained results, it is found that the LSTM model performs better than logistic regression in predicting the gender, we also compared it with machine learning models and Bayesian network and thus the hybrid approach successfully classifies the age groups. Some of these methods seem promising to be used in real-world applications in virtual assistants and healthcare.