Music genre classification using EMD and pitch based feature

Automated classification of music signal is an active area of research. It can act as the fundamental step for various applications like archival, indexing and retrieval of music data. In this work, a simple methodology is presented to categorize the music signals based on their genre. In order to capture the characteristics of the music signal of different genres, signal is first decomposed to extract the component reflecting the desired degree of local characteristics using empirical mode decomposition (EMD). Pitch based features are computed corresponding to the signal at suitable intermediate frequency range. Multi-layer perceptron network is used for classification. Experiment with GTZAN dataset and comparison with number of state-of-the-art systems reflect the effectiveness of the proposed methodology.