Feature extraction and genre-classification using customized kernel for Music information retrieval
Karthik, Savita Choudhary · Turkish Journal of Computer and Mathematics Education (TURCOMAT) · 2021
Music feature extraction and genres form a natural way to consolidate audio and they share related rhythm andtexture. We will be building a customizedfeature extraction genre classification model using customized kernel in supportvector machine that will use features representing timbre, rhythmic and pitch analysis of the audio. We train variousclassifiers like k-Nearest neighbor, Support vector machine, Logistic Regression, Neural Network on the GTZAN datasetprovided by MARYSAS. We are able to get good accuracy using Customized kernel and ensemble voting classifier andsupport vector machine on both 10-genre and 4-genre classification.