Decoding the Algorithm for Pop Music 353
fb4f writes "Over at Modplug, they have an article describing a mathematical algorithm to predict if a given song will become a hit or not. Paraphrasing the article, a Spanish company called Polyphonic HMI has made a business out of analyzing song submissions and predicting their "hitability". Here's their description of the algorithm and here's their FAQ. They claim to have predicted the commercial success of Norah Jones through this method. Here's my question (which is not fully answered in their FAQ): if they (music company executives) are currently using the algorithm to screen submissions for their "hitability", can we (people who listen to music) use the same algorithm to reject recycled tunes and encourage originality? I for one, still like the fresh talent and community feel of the tracking scene."
Not new (Score:2, Interesting)
And my question ... (Score:5, Interesting)
And here's my question: can we use this algorithm to create the hit, instead of determining wether or not it's gonna be a hit?
Classical (Score:4, Interesting)
IIRC, this was first tested on random samplings of classical music. Beethoven and Mozart scored significantly better that others.
The Manual (Score:4, Interesting)
The Manual [instrumentality.com] - how to have a number one - the easy way.
Written by the Timelords (the KLF)
(i know, this is a bit offtopic, but hey!, why not?)
Re:And my question ... (Score:2, Interesting)
it's the other way around (Score:5, Interesting)
Anyway the point is that the guy pointed out that most pop tunes were rehashes of older pop hits. Maybe you create a different style with different instruments or beat, but the underlying melody is the same. He then showed some examples, in how some modern R&B titles were lifted off some older Rock titles. It's not that hard to believe though, look at how Puff Daddy makes a living out of talking fast over music of old hits.
So in short, one way to predict if a music will be a hit is by creating a database of previous hits and test the correlation...
[and then of course, there's those who say that Classical music tried every combination possible, so nothing can be new afterwards - but that's maybe a little extreme].
Snake oil? (Score:5, Interesting)
"This software will compare the song to a database that contains the "top-30" hit songs of the past five years in order to search for mathematical similarities. The algorithm then assigns each song a score between one and 10. Any song rated more than seven is likely to become a hit."
Now think about this.. use musical eras like the 80's and early 90's as an example because it's reasonably safe to assume this technology didn't exist at that point.
Look at the charts in 5 year chunks, it all sounds the same. In the 80's, everyone either used a synthesizer or had a raging, face-melting solo at some point in the song. Or the early 90's, "grunge" was being pounded into our head incessantly.
It was like that because it was popular. Band X makes it big, and suddenly Bands X1 through X255 appear on the charts mimicing this sound. This seems to happen in, amazingly enough, cycles of 5 years.
Seems to me this software does nothing to show the "hitability" of a song, but rather telling you whether or not it sounds just like what's currently popular, and has been for the past couple years.
Seems about as magical to me as as an algrorythm claiming it can detect boys that like looking at porn.
Re:Karma Hit (Score:4, Interesting)
Re:Can't be done (Score:1, Interesting)
Re:Karma Hit (Score:3, Interesting)
Someone should make a program that when you ask it to, downloads a story (showing all comments and with the number per page at max so it doesn't have to spider and piss off slashcode) and splits the comments into -1, 0, 1, 2, 3, 4, 5 "buckets" and creates a spam filter. Then, you can apply it to other stories. Sorting by topic would be good, since saying Windows in an Apple story has a good chance of being a flame (or maybe a comparison), but saying it in a story about Longhorn [sucks] probably is less important.
Copyright (Score:2, Interesting)
Anyway the point is that the guy pointed out that most pop tunes were rehashes of older pop hits.
And this is how the situation perpetuates itself. If somebody new to the scene comes in and tries to write an original song, he will undoubtedly get bit by an earworm of a song that was popular decades ago but is still copyrighted. Then the older song's publisher will sue the rookie.
Yes, it could happen, and yes, it did happen: Bright Tunes Music v. Harrisongs Music [columbia.edu].
and then of course, there's those who say that Classical music tried every combination possible
The established songwriters' counsel will do everything in their power to downplay this theory so that they can prevail in a copyright infringement case. The argument might go a little something like this [slashdot.org]: if the rookie has heard the established songwriter's song even once on the radio, and the two songs are "substantially similar," then the rookie has infringed copyright.
Apparently, the only way [slashdot.org] to avoid a lawsuit entirely is to cover songs from the public domain.
"Melancholy Elephants" by Spider Robinson (Score:3, Interesting)
Maybe it could compose the chant melody by recognizing actual chords in the samples, and applying known hamonic rules to end up with a melody to put on top of it
But because the space of pleasing music is so littered with copyrights [slashdot.org], anybody who publishes such a song may get sued into oblivion. Believe me; I tried implementing algorithmic composition once, but I noticed bits and pieces of songs I knew to be copyrighted in its output, and who knows what other copyrights the program violated.
Along these lines, you may find this short story by Spider Robinson [baen.com] interesting.
Re:Recipe music (Score:3, Interesting)
I read that more as "our algorithm doesn't work most of the time, but if we get to rig the test, er, um, choose the musical style we want to deal with, we do ok."
If they could identify a set of interesting weights (or whatever) that their stuff comes up with, and track those weights over time, then if the time series are anywhere near smooth they could predict what will be hot next year. Then too the act of predicting the future can well change it...
Rivers Cuomo already did it. (Score:3, Interesting)
Appears to work (or at least teach him a pattern)--Weezer's damn catchy.
Re:Star success? (Score:1, Interesting)
Polyphonic HMI had nothing to do with her success, she was actually taken under the wing and helped to the top by my favorite 8-string guitarrist, Charlie Hunter (formerly of Blue Note.)
From jambase [jambase.com]: "En route to a recording session for his most recent Blue Note release, Songs from the Analog Playground, eight-string guitar wizard Charlie Hunter and his bandmates first heard the First Sessions EP when their manager was played the album way to the studio. At first note, orders ensued to find the girl at once. Contact was made and Norah was hunted down at the Midtown restaurant where she was working as a waitress. Arriving at the studio and meeting the musicians, Norah was given the lyrics to a song written by Brian Ferry entitled "More Than This." Never before seeing these words let alone knowing the song, Norah was asked to just give it a shot. After getting a feel for the song, she began to sing the band playing behind her. At song's end, Norah timidly asked, "if that was okay?" For a minute everyone in the room remained speechless, mesmerized by Norah's radiant essence. Unanimously, it was decided that yes, it was in fact okay."