Music Genre Classification: A Review of Deep-Learning and Traditional Machine-Learning Approaches
Ndiatenda Ndou, Ritesh Ajoodha, Ashwini Jadhav
Abstract
This research provides a comparative study of the genre classification performance of deep-learning and traditional machine-learning models. Furthermore, we investigate the performance of machine-learning models implemented on three-second duration features, to that of those implemented on thirty-seconds duration features. We present the categories of features utilized for automatic genre classification and implement Information Gain Ranking algorithm to determine the features most contributing to the correct classification of a music piece. Machine-learning models and Convolutional Neural Network (CNN) were then trained and tested on ten GTZAN dataset genres. The k-Nearest Neighbours (kNN) provided the best classification accuracy at 92.69% on three-seconds duration input features.