Turkish Music Genre Classification Using Convolutional Neural Network

Abstract

Music genre classification has become essential in today's world due to the explosive growth of music audio recordings. Previous research in this field primarily concentrated on exploring the classification of genres in Western music. Nevertheless, categorizing online Turkish music content remains inadequately defined, posing a challenge for the automatic classification of audio genres in the Turkish music context. The majority of earlier work on Turkish music genre classification focuses mostly on acoustic features with machine learning techniques, which limits the performance. In this research, we study automatic Turkish music genre classification with convolutional neural networks by using images obtained by the Mel-spectrogram analysis. The initial step involves the creation of a dataset comprising of the most six well-known categories of Turkish music, which include Arabesque, Pop, Rap, Rock, Turkish Classical Music, and Turkish Religious Music. Then, this research proposes a deep learning technique with convolutional neural networks and employs spectrogram images produced from Mel-Spectrogram as the input into a CNN to classify the songs into the appropriate musical genres. The proposed method has an accuracy level of about 99.62% for training and 99.80% for testing, which confirms the success of this method.