Classification of vocal and non-vocal segments in audio clips using genetic algorithm based feature selection (GAFS)
作者：
Highlights：
• A novel set of features based on the formant structure has been proposed.
• Evolutionary based feature selection has been proposed to obtain optimal feature set.
• Comparative analysis has been made with the available feature selection algorithms.
• Two datasets and four different classifiers are considered for comparative study.
• The concept of n-point moving window is introduced to improve the classification.
摘要
•A novel set of features based on the formant structure has been proposed.•Evolutionary based feature selection has been proposed to obtain optimal feature set.•Comparative analysis has been made with the available feature selection algorithms.•Two datasets and four different classifiers are considered for comparative study.•The concept of n-point moving window is introduced to improve the classification.
论文关键词：Dimensional reduction,Genetic algorithms based feature selection,Formant analysis using geometric methods,n-point moving window,Singing voice detection,Vocal and non-vocal segmentation
论文评审过程：Received 6 December 2017, Revised 8 March 2018, Accepted 4 April 2018, Available online 7 April 2018, Version of Record 13 April 2018.
论文官网地址：https://doi.org/10.1016/j.eswa.2018.04.005
