An In-Depth Analysis of Voice Gender and Its Effectiveness Using an Expert Recognition System

The paper presents an in-depth case study investigating the effectiveness of voice characteristics for gender recognition. While previous research has explored the relationship between acoustic features and gender classification, a more comprehensive understanding of the factors influencing this task is needed. The primary objective of this case study is to leverage a diverse dataset of voice samples, encompassing a wide range of speakers from different demographic and linguistic backgrounds, to identify the most salient voice features for accurate gender classification. The research seeks to gain deeper insights into the complex relationship between vocal features and gender identity. By adopting a comprehensive and multidimensional approach, this case study offers a valuable resource for researchers, speech technology developers, and practitioners in related domains, providing a deeper understanding of the effectiveness of voice for gender recognition. Our findings demonstrate that deep learning models can effectively capture the nuanced acoustic characteristics and patterns present in voice data, leading to accurate gender classification. The artificial neural network (ANN) model emerged as the top-performing architecture, achieving an impressive 98.00% accuracy and 99.76% area under the curve (AUC) score on the test set, outperforming the convolutional neural network (CNN) and long short-term memory (LSTM) models