Music Genre Classification

Introduction

A music genre is a conventional category that identifies some pieces of music as belonging to a shared tradition or set of conventions (rock, pop, metal, etc). It is to be distinguished from musical form and musical style.

Music can be divided into different genres in many different ways. The artistic nature of music means that these classifications are often subjective and controversial, and some genres may overlap. [1]

Inherited Fields

This project is based on knowledge from

Music Information Retrieval (MIR)
Audio Signal Processing for Music Applications (DSP)
Machine Learning (ML)

What Does this Repository Contain

This repository contains folders regarding:

Feature Extraction
Datasets Used
Score/Feature Scaling and Weighting
Ensemble Learning
Different Machine Learning Approaches to Improve Results

We wish to work on the most popular 8 genres as:

Blues
Country
Electronic
Hip Hop
Jazz
Metal
Pop
Rock

Dependencies

Numpy
Scipy
IPython
Scikit Learn
Librosa
ffmpeg
seaborn
pandas
Matplotlib

To work underthese you must have the following packages installed:

sudo apt-get install python-scipy python-numpy python-matplotlib ipython pip pandas ffmpeg

and the following library installed:

sudo pip install librosa seaborn

or if you work under windows:

Install Anaconda
Insert Environment Variables C:\Python27\Scripts;C:\ffmpeg\bin in Advanced System settings to support pip and ffmpeg

for more on librosa documentation: http://librosa.github.io/librosa/

References