Hit Song Science

[Thulsa Doom]

Have you heard about this? This stuff called Polyphonic HMI software has been developed to help determine the liklihood of any given song becoming a billboard hit. Say what? Yes. We have reached the point in our technology where software is sophisticated enough to analyze music (the structure, rhythms, changes in keys) convert all this to raw data and compare the raw data to a known data base of "hits" and thereby figure out if its likely to be succesful at all. Apparently this software predicted the massive popularity of Nora Jones recent album which was very unlikely intuitively. I mean teenie boppers buying a young lounge jazz crooner by the millions? come on. but the software said: multipe hits. and it got it right. woah.

So this means that all music will now be essentialy formulated (at least culled out) by computer. Big labels are already using this thing on all their prospects. What will this mean for the art of music making in the next half century?

heres the premise:

Much of what attracts us to a particular song is found in the basic structure of the music. Particular rhythms, changes in key and certain melodic patterns define the psychological and very human response we all have to music.

Polyphonic HMI has developed proprietary music analysis technologies capable of identifying music preferences of a user or the whole current recorded music market and intelligently selecting music to recommend to the user or to release as a single.

Our technology not only allows similarities to be identified between existing successful music and unreleased or unsigned music, but can also use that data to identify emerging trends as the music landscape changes.

 

[HeXp£Øi±]

Very interesting. I'd love to get my hands on that software and test out my own beats.

Do you have a link to that article?

 

[Ms Ann Thrope]

My initial reaction was horror, Thulsa. But after thinking about it for a few minutes it occured to me that very little will change. Record labels along with movie and television production studios have been "cloning" successful bands and narratives for a long time now. So now they have a little more control over the process maybe? They always had a lot of control.... they held the keys to the bank vault. The creative process is already controlled by our markets. Economic forces shape the type of music, art and literature that is consumed by the mass market.

This doesn't mean that artists won't continue to compose, paint, write independent thoughts. They're not rewarded NOW, and it hasn't stopped the flow. As a consumer, it may get a little harder for me to find their products, but I will.

and by the way.... do ANY serious jazz lovers like Norah Jones?

 

[Squiggy]

Computers are also allowing the independents to produce much better stuff at home. And the internet has proven a tremendous marketing asset for them. It should all pan out in the long run. A successful independent whose work was software rejected will eventually catch the attention of the lable's human elements and be sucked into the void...

 

[Bungi]

and by the way.... do ANY serious jazz lovers like Norah Jones?

I'm not a serious jazz lover, but I don't like her anyway.

 

[chcr]

My initial reaction was horror, Thulsa. But after thinking about it for a few minutes it occured to me that very little will change.

Yep. Labels have been doing the formula thing for years. Douglas Adams wrote a book once where a guy became very rich by making software that after you put in the data and the conclusion you wanted, would then come up with a series of plausible seeming steps to support the conclusion (bought up lock stock and barrel by the US military). This has always made me think of A & R guys.

and by the way.... do ANY serious jazz lovers like Norah Jones?

Evidently there are some, but I don't know any of them.

 

[Camelyn]

I saw a piece about this software on TV recently. The interviewer raised the same kinds of concerns. Apparently this software does not analyse the "sound" of the musical piece, but the structure. The creators of this software analysed "Hit songs" From Bethoven to Brittany Spears, and from nursery rhymes to rap music. The hits all have a balance of structure that can be plotted on a type of graph. The grap has "hit custers" and if the piece of music falls into those clusters, it has the potential of being a hit. It doesn't matter if it's a new pop song or a piece beaten out on pots and pans.
What I got from it was that it doesn't mean that all hit music must sound the same.

Currently, musicians uses the software to deternime which tracks to release as singles off their albums. Up-and -commers are also getting their pieces analysed and presenting the results to record conpanies in the hopes of improving their chances at a deal.

The company that created the software and does the anaysis works independantly of the record industry.

 

[Thulsa Doom]

Yeah thats what I found most amazing. That its not just following a "formula" like industy execs have been doing. Its strict detailed structural analysis. They said that in this system U2 and Beathoven had very similar read outs.

and by the way.... do ANY serious jazz lovers like Norah Jones?

I dont mind her. I think its because she is a massive breath of fresh air compared to the usual crap. i mean do you really lump her in as exactly the same as Britney Spears and that lot? I think its fantastic if a 13 year old girl gets into music by way of Norah Jones rather then Britany Spears.

 

[Ms Ann Thrope]

I dont mind her. I think its because she is a massive breath of fresh air compared to the usual crap. i mean do you really lump her in as exactly the same as Britney Spears and that lot? I think its fantastic if a 13 year old girl gets into music by way of Norah Jones rather then Britany Spears.

Good point.... but there are SO many other brilliant jazz vocalists, I can't help but feel a little disappointed.... not surprised, though.