Ars Technica Newsdesk
Posted 4/10/2003 - 1:30AM, by
Fred "zAmboni" Locklear
Researchers have previously used compression algorithms to determine the language and authors
of text documents. Now, a group at the Dutch National Research Institute
are using these techniques on music files to see if they can classify musical genres and identify artists. They
subtracted any data unrelated to the actual music, such as digital ID tags,
to create a data string representing only the rhythm and melody of the tune.
Using a program called Bzip2, they followed a similar procedure as with the
text files, measuring how similar each piece was to every other. Then they
plotted the results in a way that produces a tree-shaped pattern, in which
similar pieces cluster together on the same branch. In a test with
12 each of jazz, classical and rock pieces, the results were fairly good.
Ten of the jazz, nine of the rock and most of the classical pieces ended
up in three distinct branches of the tree.
When the technique
was applied to music from within a genre, it was able to cluster different
composers on separate branches. It may be possible to identify anonymous
and unknown composers with this system along with how much one composer contributed
to a collaboration.
[Discussion | Send to a Friend]
|