Applying Compressive Sensing techniques in an effort to separate several speaking voices with minimal effort.
Contributors: Eli Saracino (esaracin@bu.edu) Andy Huynh (ndhuynh@bu.edu)
##code/ This directory contains the sound data used to test our reconstructions, as well as the code used to perform the signal separation. Data used includes a mix of Clint Eastwood and Graham talking over each other, as well as several samples taken from the Interspeech 2006 source separation competition.
##paper/ Contains our project's written report as a .tex file and a .pdf file. Any plots written by the MATLAB script in the /code directory that were used in this report are also in this directory, as are any papers references in our report.
##slide/ Contains the .pptx file used to present the state of our project in early December, 2017.