Vocal source builds divergence in gender recognition.

Tholeti, Guru Sree Ram; Ghanta, Deepika; Chilukuri, N.V.S. Guru Sai Sarma; Bano, Shahana

Abstract

The purpose of this study was to investigate whether artificial intelligence could be used to determine a person's gender based on the sound of their voice. The research examined different machine learning and deep learning algorithms for gender classification based on voice. The study used multi-layer perceptron (MLP), random forest, decision tree and logistic regression models, and compared their performance. MLP was shown to achieve an accuracy of 96.84%; random forest achieved 96.42%; decision tree achieved 96.21%; and logistic regression achieved 89.37%.

Citation

THOLETI, G.S.R., GHANTA, D., CHILUKURI, N.V.S.G.S.S. and BANO, S. 2022. Vocal source builds divergence in gender recognition. In Kumar, A., Senatore, S. and Gunjan, V.K. (eds.) Proceedings of the 2nd International conference on data science, machine learning and applications (ICDSMLA 2020), 21-22 November 2020, Pune, India. Lecture notes in electrical engineering, 783. Singapore: Springer [online], pages 171-183. Available from: https://doi.org/10.1007/978-981-16-3690-5_16

• Conference Name: 2nd International conference on data science, machine learning and applications (ICDSMLA 2020)
• Conference Location: Pune, India
• Start Date: Nov 21, 2020
• End Date: Nov 22, 2020
• Acceptance Date: Oct 15, 2020
• Online Publication Date: Nov 9, 2021
• Publication Date: Dec 31, 2022
• Deposit Date: Jun 6, 2024
• Publicly Available Date: Jun 6, 2024
• Publisher: Springer
• Pages: 171-183
• Series Title: Lecture notes in electrical engineering
• Series Number: 783
• Series ISSN: 1876-1100; 1876-1119
• ISBN: 9789811636899
• DOI: https://doi.org/10.1007/978-981-16-3690-5_16
• Keywords: Speech recognition; Vocal register classification; Machine learning; Random forests; Decision trees
• Public URL: https://rgu-repository.worktribe.com/output/2063949