Multiple voice disorders in the same individual: Investigating handcrafted features, multi-label classification algorithms, and base-learners

Sylvio Barbon, Rodrigo Capobianco Guido, Gabriel Jonas Aguiar, Everton José Santana, Mario Lemes Proença Jr., Hemant A. Patil

Abstract: Highlights•The proposed approach detects multiple voice disorders in one individual.•Energy, zero-crossing rates, and entropy were successfully used as features.•Over 90% of accuracy was obtained under SVD database using SMOTE.