"Hit Song Science"

[bradm]

There's a story on Slashdot today, which links to a story in the Guardian newspaper in the UK about a new technology called "Hit Song Science". It's an artifical intelligence program that

isolate and separate 20 aspects of song construction including melody, harmony, chord progression, beat, tempo and pitch and identifies and maps recurrent patterns in a song, before matching it against a database containing 30 years' worth of Billboard hit singles - 3.5m tunes in all. The program then accords the song a score, which registers, in effect, the likelihood of it being a chart success.

It's done by a Spanish company called Polyphonic HMI.

I'm not sure whether this is a good thing or a bad thing. While it might lead to bland, mechanized, music without personality, it has predicted the success of artists like Norah Jones, and since my Mom likes her music, I guess some good came out of it...

Aloha,

Brad

[Booche]

Sharon or Catphish's reply in 3......2.........1

[Im going home Donny]

I love Nora Jones version of crazy..its beautifull.

She's a cool lady.

[Dynamo]

I've heard of this before...and judging by what's out there on the charts these days, a lot of silly producers are using it!

[Boochawan]

The whole concept of a "hit song" is an interesting one. It's incredibly multi-factorial. On one hand, if a hit song can be created by the suggested formula of this 'science,' then wouldn't everyone be using it, let alone have caught on years ago? Secondly, if local band X wrote and performed a song that followed this paradigm there is no gaurantee that the song would become a "hit" (as defined by the popular industry) because of the limited population that the local band is exposed to. This pseudoscience merely indicates the likelihood, a probability, and not an exact truth. Given that, only an increased probability of a hit is likely if the song, and the performer have the wide-spread exposure to a large enough audience - and confired by the authoritative music industry, e.g. MTV - only then to deem it a "hit."

[ggrtrhhrtgg]

I would think music is more of an art than a science.

[snarfmaster C]

I wonder how a 17 minute Fraklin's Tower would rank