Re: Solo ea


Subject: Re: Solo ea
From: Eliot Handelman (eliot@generation.net)
Date: Sun Feb 13 2005 - 22:13:45 EST


Richard Wentk wrote:

> At 07:06 13/02/2005, you wrote:
>
>> What about Hit Song Science? Company wrote program that clusters
>> hits, using some sort of
>> analysis: for 50 bucks, you send in your tune and they rate it by
>> cluster proximity. Some musicians
>> seem to think this works.
>
>
> Is that intellectual? I wouldn't it call it intellectual in the
> academic sense. It looks more some rather clever statistics, perhaps
> with a certain amount of clandestine human talent spotting.

The same stuff is going on in academe -- there's the princeton project
(forget guy's name) to automatically classify
music and find "similar" music, Dannenberg & Vercoe have also autrhored
classification papers and I'd judge it to
be the dominant idea about how music information retrieval ought to
work. The company's algorithms are probably
related to everyone else's with hidden markov processes etc. So I don't
see much discontinuity here.

That the project represents total music-intellectual bankruptcy goes
without saying.

-- eliot



This archive was generated by hypermail 2b27 : Sat Dec 22 2007 - 01:46:06 EST