Music Genre Classification Using Adam Algorithm of Convolutional Neural Network

Even though technology has been evolving rapidly lately, music classification is still definitely a major task in the Music Information Retrieval (MIR) domain. Music genre classification is a key challenge in Music Information Retrieval (MIR), aiming to identify the genre, style, and mood of audio tracks. This study explores the use of Convolutional Neural Networks (CNNs) with the Adam optimizer for music genre classification. We conducted experiments to evaluate the performance of our proposed model, which incorporates advanced machine learning techniques to improve classification accuracy. Our approach involves extracting features from audio files, converting them into Mel spectrograms, and training the CNN model using Python. The results demonstrate a high classification accuracy of 98.5%, significantly improving upon previous methods. Additionally, GPU acceleration enhanced the training speed by five times. Future work includes developing a mobile application for real-time classification and exploring integration with speech recognition technologies