Improve README, add whitespace to match original, fix some broken characters

[endolith]

Improve README citation

• Include more details about original paper, and a link to a downloadable PDF.
• Move "this website" under course information.
• Clarify what CASA and DUET stand for.
• The files have been modified somewhat from the original, not a literal copy.
• Clarify audio files source and DUET variant

Fix whitespace in .m files

Added spaces and newlines to match the appearance in DuetSourceSeparationTutorial.pdf more accurately. The lack of spacing had caused some words to be mashed together.

Fix special characters in .m files

The PDF formatting converted hyphens to minus signs, carets to circumflex accents, asterisks to asterisk operators, and underscores to blanks, which resulted in some mangled text and variable name differences.

[endolith]

I also tried to put your comments in [square brackets] to separate them from the original comments, but I'm not sure what to do about the numbered items, like this:

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%    step 4,5,6,7
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% 4. peak centers (determined from histogram)  [THIS IS DONE BY HUMAN.]
% [
% 4) Determine how many peaks there are in the histogram.
% 5) Find the location of each peak.
% ]

It's confusing that there are multiple step 4, multiple step 6, etc.

[endolith]

Also in the readme I said the WAV files are not the same as the ones in the paper (because the waveforms shown don't match)

But do you remember where you got them? Are they from this?

For a demonstration of degenerate demixing (5 sources, 2 mixtures), listen to the following:

Five Source TIMIT Synthetic Mixtures

These two mixtures were created using an anechoic mixing model on sources taken from the from the TIMIT voice file database. Mixtures: x1.wav|x2.wav DUET: u1.wav|u2.wav|u3.wav|u4.wav|u5.wav

I can't find any other copies of these files in Wayback Machine, only in your repo, so I can't compare to confirm that they're the same files.