MUSIC GENRE CLASSIFICATION USING MACHINE LEARNING

Automatic music genre classification has become increasingly important for digital music management, recommendation systems, streaming platforms, and automated tagging. This work presents a deep learning–driven framework that classifies music genres using Mel-Frequency Cepstral Coefficients (MFCCs) along with complementary spectral–temporal features such as Root Mean Square (RMS) energy, Spectral Centroid, Tempo (BPM), and Zero Crossing Rate (ZCR). Convolutional Neural Networks (CNN), LSTMs, and hybrid CNN–LSTM architectures are employed to learn high-level audio representations. The proposed system demonstrates improved accuracy and scalability compared to conventional machine-learning approaches and reduces the dependency on manual feature engineering.