Slashdot Mirror

I know that hit music science may be something that people have heard of, but the whole technology of the measuring and subsequent optimizing of music output scares me. We must be aware of it. This has been around for about 4-6 years now people. No wonder you think indie music is better. http://www.polyphonichmi.com/technology.html

There's a company in Barcelona called Polyphonic HMI that have a service called Hit Song Science. It's been around for several years and they already work for most of the major labels and top producers. http://www.polyphonichmi.com/

To be fair, those two stories (that same time-shifted story?) are covering this product, while the current story is just a random Media Lab project.

(I very loosely know the guy who was working on this -- he graduated last year. His algorithm wasn't that successful...)

In a nutshell, they took the songs from the Billboard charts from the past x years and ran them through a computer looking for dozens of different values (key, song length, tempo, mix of music to lyrics, etc). Because they're basing their statistics on songs that already were popular, they're guessing that all songs that have xx, yy and zz values for specific criteria also has the potential to be a hit.

From the article.

In a retail environment, both the "more like this" and the "music taste test" can be efficiently presented on in-store terminals, or on a retail website. The same technique could be applied to many other situations, such as automatically recommending songs from a personal collection as a playlist, or anywhere that commonalities between pieces of music can be useful.

Now, call me a cynic but i see a very high possibility for corruption in the retail market. When the public gets a taste for their product, the recording industry can just use it for more marketing. All they have to do is insert what they want the public to think is going to be "the next big thing" and someone will bite, causing a trend. Maybe I should find my tinfoil hat?