This repository contains all the scripts required for feature extraction on audio files and speech/music classification using MATLAB's Fuzzy Logic Toolbox
Music and Speech files should be placed in the MusicSpeech subfolder
The script demoFull.m executes all the scripts required for a complete demonstration of the system
A complete summary of the results of all tested predicition models (i.e. Confusion Matrices, Precision & Recall Scores and others) is included in the folders MATLABOutput and WekaOutput in the form of MATLAB workspace .mat files and .txt files of Weka's GUI output.
Musan dataset files have been picked randomly as a part of the full dataset, the chosen files can be found at: https://drive.google.com/open?id=1xtB7YP_lND3BDYUyvfT5G-RH79j06ZkF