Impact of Machine Learning in Music Genre Classification using CNN
Jorige Venkatesh, Karthik Kannan, M. Ayyadurai, M. Sathish · 2023
The rapid expansion of available music tracks has made the classification of music genres increasingly crucial. To achieve accurate genre predictions, machine learning techniques have been widely employed and proven highly efficient. In this study, we focused on enhancing the accuracy of music genre classification systems by training a Convolutional Neural Network (CNN) model using Keras. Music genre classification serves as a valuable tool for music learning, offering a reliable method of categorization. Given the immense amount of data and the inherent complexity in identifying music genres, we leveraged the power of deep neural networks to classify ten distinct genres. To evaluate the effectiveness of our approach, we compared the performance of the CNN model to other existing models using the GITZAN dataset. Our results demonstrated the success of the CNN model in accurately classifying music genres, surpassing previous studies while requiring less data. This research contributes to the field of music genre classification by providing an advanced and efficient solution using deep learning techniques.