Index of /~echew/papers/CMMR2004

      Name                    Last modified       Size  Description


[DIR] Parent Directory        13-Sep-2009 18:19      -  
[   ] CMMR-20040527.pdf       07-Mar-2005 01:47   350k  
[   ] ecxw-cmmr2004.pdf       23-Jun-2004 23:51   659k  
[TXT] reference.txt           25-May-2007 05:00     1k  

In this directory are the PDF files for the paper and presentation titled

"Separating Voices in Polyphonic Music: A Contig Mapping Approach"
by Elaine Chew and Xiaodan Wu ({echew,xiaodanw}@usc.edu)

This paper was presented at the 
2nd Computer Music Modeling and Retrieval Symposium
in Esbjerg, Denmark.  26-29 May, 2004.

The paper was published in the
Proceedings of the 2nd Computer Music Modeling and Retrieval Symposium
Esbjerg, Denmark, May 26-29, 2004.

The revised paper (this draft) will be published in 
Lecture Notes in Computer Science #2771, Springer Verlag: Berlin.
Proceedings of the 2nd Computer Music Modeling and Retrieval Symposium
Click on reference.txt for the BibTeX reference.

The conference website is at
http://www.cs.aue.auc.dk/cmmr/cmmr2004

THE COMPLETE PAPER, text with figures, can be viewed as a PDF document.
Click on ecxw-cmmr2004.pdf if you wish to view the paper in PDF format.

THE CONFERENCE PRESENTATION can be viewed as a PDF document.
Click on CMMR-20040527.pdf if you wish to view the presentation in PDF format.

--'--,--'--,--'--,--'--,--'--,--'--,--'--,--'--,--'--,--'--,--'--,--'--

"Separating Voices in Polyphonic Music: A Contig Mapping Approach"
by Elaine Chew and Xiaodan Wu ({echew,xiaodanw}@usc.edu)

ABSTRACT.  Voice separation is a critical component of music
information retrieval, music analysis and automated transcription
systems. We present a contig mapping approach to voice separation
based on perceptual principles. The algorithm runs in O(n2) time, uses
only pitch height and event boundaries, and requires no user-defined
parameters.  The method segments a piece into contigs according to
voice count, then reconnects fragments in adjacent contigs using a
shortest distance strategy.  The order of connection is by distance
from maximal voice contigs, where the voice ordering is known. This
contig-mapping algorithm has been implemented in VoSA, a Java-based
voice separation analyzer software.  The algorithm performed well when
applied to J. S. Bach's Two- and Three-Part Inventions and the
forty-eight Fugues from the Well-Tempered Clavier. We report an
overall average fragment consistency of 99.75%, correct fragment
connection rate of 94.50% and average voice consistency of 88.98%,
metrics which we propose to measure voice separation performance.