Abstract

Music is like a mirror, and it tells people a lot about who you are and what you care about, whether you like it or not. Music can be classified into taxonomies based on genre, performer, composer or geographic or cultural point of origin. Music genres can be seen as categorical descriptions used to segregate music based on various characteristics such as instrumentation, pitch, rhythmic structure, and harmonic contents. The top 10 genres in the music industry are blues, classical, country, disco, hip-hop, jazz, reggae, rock, metal and pop. Automatic music classification is an area of research that has been receiving a great deal of attention in recent years due to the rapid growth of digital entertainment industry. There are two major challenges with music genre classification: Firstly, musical genres are loosely defined, so that people often argue over the genre of a song. Secondly, extracting differentiating features from audio data that could be fed to the model is a nontrivial task. Although music genre classification has been a challenging task in the field of Music Information Retrieval (MIR), automatic music genre classification is important for music retrieval in large music collections on the web. This project aims to build a machine learning classifier after scrutinizing various machine learning algorithms that classifies music based on its genres. The chosen classifier, Support Vector Machines then learns from the data, explores the performance of various features extracted from the audio signal and classifies the genre of the audio input. This project can be extended to develop various systems like music genre-based disco lights and emotion-mapped music systems.

Description