FREE counter and Web statistics from sitetracker.com
collision detection
content | discontent
send me yours
December 17, 2003
"Hit Song Science": another of my NYT mag pieces

Here's yet one more of my essays in this week's New York Times Magazine's "Year in Ideas" issue:

Hit Song Science

When Norah Jones released her first album, she was a long shot at best. ''Come Away With Me'' was filled with mellow, sultry tunes -- precisely the opposite of the histrionic diva pop crowding the charts. Virtually no one expected Jones to score a major hit.

No one, that is, except for a piece of artificial intelligence called Hit Song Science, a program that tries to determine, with mathematical precision, whether a song is going to be a Top 40 hit. When the scientists fed Jones's album into that computer, alarm bells went off: the program predicted that eight tracks would hit the charts. ''We were like, whoa, that's funky,'' says Mike McCready, the C.E.O. of Polyphonic HMI, the Barcelona-based company that developed the software application. A few months later, Jones's album went multiplatinum -- and Hit Song Science had proved it could pick a hit as well as Clive Davis.

But how? At the heart of the program is a ''clustering'' algorithm that locates acoustic similarities between songs, like common bits of rhythm, harmonies or keys. The software takes a new tune and compares it with the mathematical signatures of the last 30 years of Top 40 hits. The closer the song is to ''a hit cluster,'' the more likely -- in theory -- that the kids won't be able to resist it. Yet the weird thing is, songs that are mathematically similar don't necessarily sound the same. The scientists found that U2 is similar to Beethoven, and that Van Halen shares qualities with the piano rock of Vanessa Carlton. Even more bizarrely, 50 Cent's throbbing rap tune ''If I Can't'' correlates with ''(There's) No Gettin' Over Me,'' a twangy country ditty by Ronnie Milsap.

This year, several record companies began using Hit Song Science to help pick which songs on an album to promote. Others are now using it in the studio, taking a rough mix of a new song, checking to see how hit-worthy it is, then tweaking it until it has ''good mathematics,'' as McCready puts it. He can foresee a day when most major hits will have been vetted by algorithms.

Which is, depending on how you look at it, either a wonderful breakthrough for science or an incredibly bleak statement about the music industry. Critics for years have complained that record labels produce only bland albums that mimic what's already popular. But Hit Song Science takes that trend to its logical absurdity: it does not merely aim at the middle of the road -- it calculates it, with scientific precision. -- Clive Thompson

Posted by Clive Thompson at December 17, 2003 02:18 PM

Trackback Pings

TrackBack URL for this entry: http://www.collisiondetection.net/mt3/mt-tb.cgi/656

Listed below are links to weblogs that reference "Hit Song Science": another of my NYT mag pieces:

» maths and melodies from BARISTA
You can't pick a successful pop song right? Here is an extremely irritating idea: "a piece of artificial intelligence called Hit Song Science, a program that tries to determine, with mathematical precision, whether a song is going to be a... [Read More]

Tracked on January 1, 2004 9:08 AM

» online poker from online poker
You may find it interesting to visit some information dedicated to party poker online poker texas hold em [Read More]

Tracked on March 18, 2005 1:26 AM

Comments

Clive,

Is the mathematical signature of popular songs a universal concept? What happens if you feed, say, bhangra into HSS? Are there mathematical song traits that are inherently appealing to humanity as a whole; does that mean there's an unconscious musical aesthetic ingrained in the human brain? They could do some fascinating evo-psych-style basic research with this program. Maybe even see if there's any difference in fMRI results between hit songs and B-sides.

Okay, now I'm sounding completely ridiculous. Too much coffee.

Posted by: Mary at December 17, 2003 2:40 PM

Ahahaha! No, those are really interesting questions. I think -- though I'm not sure -- that the A.I. guys who created Hit Song Science stuck mostly to Western pop music and Western classical music. But don't quote me on that; I'm not sure.

As to whether there are hardwired aspects to our musical aesthetic ... there are plenty of scientists who are investigating precisely that point! I recently met the neuroscientist Justine Meaux, whose Phd thesis was on rhythm -- how and why our brains are able to to hear a musical rhythm and tap along with it. No other animal can do that: Not chimps, not apes, not birds. She did a bunch of fMRI studies and was trying to work on precisely that question.

Posted by: Clive at December 22, 2003 12:47 PM

okay, the hit song science people are incredibly vague with what exactly there doing. everyone's impressed because they wave around some fancy lingo about "mathematical patterns" and "algorithms" but the way they analyze these patterns is crucial because there are so many different ways to look at data. isn't anyone else skeptical?

Posted by: hello at December 25, 2003 10:06 PM

Well, the way HSS works is a two-step process. The first step is the decomposition of a song into its components -- i.e. the beat, the key, and about 20 or 30 other data points. The folks who make HSS won't tell anyone, including journalists like me, how that works; it's their Pepsi formula, as it were. So if one wanted to be skeptical, that'd be the place to do it -- because it's impossible to assess the validity of a technique the premises of which the inventors won't explain.

The second step of the HSS process is simpler: The software takes the decomposed song and checks to see how closely it "clusters" to other known clusters of hit songs. Clustering algorithms are pretty common in artificial intelligence, so I can believe that this part works pretty much as described.

But by all means, it's definitely worth casting a skeptical eye on any new invention -- particularly a secretive one!

Posted by: Clive at December 26, 2003 10:38 PM

This is nice to know. As of January 1st, Mike McCready is our manager. We are a Female fronted high energy rock band and we are based out of Tempe,AZ. We have built relations with FENDER and will be in the 2004 Fender Catalog.
We have recently been nominated for 3 awards by Badstain Records (PHX, AZ):

-Best Female
-Best New Song (101)
-Best New Artist

You can vote by email

skapunkawards@badstainrecords.com

The artical, nice work!

Keep your eye out for us...

www.girlkicksboyonline.com

Sincerely,
Stacey Funk

Posted by: stacey funk at January 14, 2004 6:40 PM

This is nice to know. As of January 1st, Mike McCready is our manager. We are a Female fronted high energy rock band and we are based out of Tempe,AZ. We have built relations with FENDER and will be in the 2004 Fender Catalog.
We have recently been nominated for 3 awards by Badstain Records (PHX, AZ):

-Best Female
-Best New Song (101)
-Best New Artist

You can vote by email

skapunkawards@badstainrecords.com

The artical, nice work!

Keep your eye out for us...

www.girlkicksboyonline.com

Sincerely,
Stacey Funk

ps Mary needs to read the artical a little closer so that she understands the concept...
Someone GET MARY ANOTHER CUP.

Posted by: stacey funk at January 14, 2004 6:44 PM

Cool -- great to hear from you! Good luck with everything! Great web site too, BTW.

Posted by: Clive at January 14, 2004 7:00 PM

Nice site. thx.

Posted by: Online Casino at January 16, 2004 7:47 PM

When a variable is finished with it's work, it does not go into retirement, and it is never mentioned again. Variables simply cease to exist, and the thirty-two bits of data that they held is released, so that some other variable may later use them.

Posted by: Cornelius at January 19, 2004 10:05 PM

This code should compile and run just fine, and you should see no changes in how the program works. So why did we do all of that?

Posted by: Dudley at January 19, 2004 10:05 PM

The most basic duality that exists with variables is how the programmer sees them in a totally different way than the computer does. When you're typing away in Project Builder, your variables are normal words smashed together, like software titles from the 80s. You deal with them on this level, moving them around and passing them back and forth.

Posted by: Drugo at January 19, 2004 10:05 PM

For this program, it was a bit of overkill. It's a lot of overkill, actually. There's usually no need to store integers in the Heap, unless you're making a whole lot of them. But even in this simpler form, it gives us a little bit more flexibility than we had before, in that we can create and destroy variables as we need, without having to worry about the Stack. It also demonstrates a new variable type, the pointer, which you will use extensively throughout your programming. And it is a pattern that is ubiquitous in Cocoa, so it is a pattern you will need to understand, even though Cocoa makes it much more transparent than it is here.

Posted by: Hercules at January 19, 2004 10:06 PM

This back and forth is an important concept to understand in C programming, especially on the Mac's RISC architecture. Almost every variable you work with can be represented in 32 bits of memory: thirty-two 1s and 0s define the data that a simple variable can hold. There are exceptions, like on the new 64-bit G5s and in the 128-bit world of AltiVec

Posted by: Emmanuel at January 19, 2004 10:06 PM

For this program, it was a bit of overkill. It's a lot of overkill, actually. There's usually no need to store integers in the Heap, unless you're making a whole lot of them. But even in this simpler form, it gives us a little bit more flexibility than we had before, in that we can create and destroy variables as we need, without having to worry about the Stack. It also demonstrates a new variable type, the pointer, which you will use extensively throughout your programming. And it is a pattern that is ubiquitous in Cocoa, so it is a pattern you will need to understand, even though Cocoa makes it much more transparent than it is here.

Posted by: Josias at January 19, 2004 10:07 PM

We can see an example of this in our code we've written so far. In each function's block, we declare variables that hold our data. When each function ends, the variables within are disposed of, and the space they were using is given back to the computer to use. The variables live in the blocks of conditionals and loops we write, but they don't cascade into functions we call, because those aren't sub-blocks, but different sections of code entirely. Every variable we've written has a well-defined lifetime of one function.

Posted by: David at January 19, 2004 10:07 PM

We can see an example of this in our code we've written so far. In each function's block, we declare variables that hold our data. When each function ends, the variables within are disposed of, and the space they were using is given back to the computer to use. The variables live in the blocks of conditionals and loops we write, but they don't cascade into functions we call, because those aren't sub-blocks, but different sections of code entirely. Every variable we've written has a well-defined lifetime of one function.

Posted by: Rook at January 19, 2004 10:07 PM

This will allow us to use a few functions we didn't have access to before. These lines are still a mystery for now, but we'll explain them soon. Now we'll start working within the main function, where favoriteNumber is declared and used. The first thing we need to do is change how we declare the variable. Instead of

Posted by: Fulk at January 19, 2004 10:08 PM

But variables get one benefit people do not

Posted by: Dionise at January 19, 2004 10:08 PM


  • カリビアンコム caribbeancom

  • カリビアン caribbean

  • 米国性動画通信 2345 2345.tv 外人 海外 洋物 洋もの ポルノ 無修正ビデオ

  • オナニーコム 0721.com 0721

  • コスチュームシアター costumetheater

  • DXLIVE ライブチャット

  • DXライブ デラックスライブ ビデオチャット

  • EXSHOT 動画チャット

  • EXショット アダルトチャット

  • Girls On Air GirlsOnAir

  • ガールズオンエアー GOA

  • 一本道 一本堂 1本道

  • HGMO HGMO H:G:M:O

  • 東京真夜中DX TOKYONIGHTS

  • URAYA URAYAOnlineTV うらや オンライン TV

  • X-GALLERY XGALLERY Xギャラリー

  • 日本人のおしっこEX JapanesePeeEX Japanese Pee EX

  • 赤外線盗撮の世界 XRAY

  • Erox EroxJapan Z EroxJapanZ エロックス エロックスジャパンZ

  • URAMOVIE 裏ムービー

  • HYPER裏ビデオ通信 ハイパー裏ビデオ通信 PINKEYES.COM

  • 出会い PURE21

  • エロアニメ TV エロアニメTV EroanimeTV Eroanime TV

  • ネットコミック NETCOMIC

  • 週間ドラムカン 文庫ドラムカン

  • JapaBeauty JapaBeauty.tv ジャパビューティー

  • eroika eroika.com エロイカ

  • 抜天市場 抜き天 抜店 nukiten

  • 画像

  • 動画

  • アイコラ

  • 盗撮

  • 無料

  • 写真集

  • 無修正

  • 覗き のぞき ノゾキ

  • サンプル SAMPLE

  • 壁紙

  • レイプ

  • パンチラ ぱんちら

  • コスプレ

  • 裏ビデオ

  • アダルト あだると

  • ダウンロード DOWNLOAD

  • AV女優タレントアイドル

  • 巨乳

  • ブルマ ぶるま

  • セーラー服 制服 ブルセラ

  • マンコ まんこ

  • セックス SEX

  • エッチ えっち

  • 熟女 人妻

  • オナニー おなにー

  • 女子校生 女子高生

  • エロ えろ ero

  • ヌード ヘアヌード nude

  • 99 BB 9BB 99B 99bb 99bb.com GON 裏ビデオ 無修正 有料 会員制

  • フリーセックス フリーセックスジャパン FreeSexJapan Free Sex Japan

  • 東京キュートガールズ TokyoCuteGirls Tokyo Cute Girls

  • 覗き屋 のぞきや のぞき屋 Nozokiaya

  • 直撃ドットコム Chokugeki.com

  • 熟まん 熟マン Jyukuman

  • Jハードコア Jhardcore J hardcore

  • JPムービーズ JPmovies JP movies

  • プリティーピンク プリティーピンクジャパン PrettyPinkJapan Pretty Pink Japan

  • JPチックス JPChicks

  • ハードポーン ハードポーンジャパン HardPornJapan Hard Porn Japan

  • J-SMUT JSMUT Jスマット

  • ヌキヌキ学園 ぬきぬき学園

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • .

  • Posted by: julia at January 24, 2004 6:56 PM

    why not je suis un chien Arf ! Arf ! c'est le rap qu'il faut bande de baltringues pourquoi vous faites les narco?

    Posted by: Big D at August 9, 2004 3:12 AM

    SEXO EN LO PRADO

    Posted by: GIGOLO CHILE at December 28, 2004 7:43 PM

    you should have some hit songs of thi year

    Posted by: rinal at January 9, 2005 4:57 AM

    Post a comment

    Thanks for signing in, . Now you can comment. (sign out)

    NOTE: If you posted a comment and you can't see it -- try refreshing your browser.


    Remember me?