EnVGG16: Enhanced Deep Learning Model for Music Genre Classification

In recent years, deep learning has revolutionized numerous domains, offering unprecedented performance in tasks ranging from computer vision to natural language processing. One such area witnessing significant advancements is music genre classification, a crucial problem in music information retrieval with wide-ranging applications in recommendation systems, streaming platforms, and digital archiving. Accurate genre classification enhances user experience and facilitates efficient content organization. Traditional machine learning approaches often struggle with the complexity and richness of audio signals. Deep learning, particularly convolutional neural networks (CNNs), has emerged as a powerful tool for learning abstract and hierarchical representations from raw or transformed audio data. This paper explores the application of an enhanced VGG16 architecture for music genre classification. VGG16, originally developed for image classification, is adapted to process spectrograms of audio tracks, leveraging its deep feature extraction capabilities. Our enhancements focus on architectural modifications and training optimizations, aiming to improve classification accuracy and robustness. Experimental results demonstrate the efficacy of the proposed approach compared to baseline models. The study underscores the potential of transfer learning and architectural adaptation in advancing deep learning applications for audio analysis tasks.