Re: Classifying music samples (MP3/WAV/...)

robin nospam at acm.org
Fri, 24 Sep 1999 11:57:12 +0100

On Wed, 22 Sep 1999 18:32:50 +0100, Paul Ashton <paul666 nospam at mailandnews.com> wrote:
> It seems to me that there has to be a better way to classify music
> than to rely on a CD TOC. How hard would it be to perform
> some sound analysis of the underlying waveform, extract some
> fundemental characteristics and then use those as the key
> to a database? Granted, it's not going to be very simple, but
> is a lot simpler than, say, voice recognition.
Should be quite simple. The main problem for use as a disk identification
mechanism would be the time it takes to sample and process the audio.

On Wed, 22 Sep 1999 13:20:38 -0400, Rob Reid <reid nospam at astro.utoronto.ca> wrote:
> This is a great idea, but as is it wouldn't distinguish between different
> compilations with the same first song. But a hash of the first few bars of
> each song on the CD should do the trick.
It's is a track identification mechanism, rather than one for disks.
I think that's good because it copes with the common occurrence of
the same track appearing on multiple compilations. To identify the
compilation as a whole, should that be needed, the system would just
have to examine as many tracks as it takes to disambiguate the disk.

Robin.

-- 
R.M.O'Leary <robin nospam at acm.org> +44 7010 7070 44, PO Box 20, Swansea SA2 8YB, UK