Pitchfork on music analysis and discovery software
Pitchfork's excellent (also rather long. Grab your hot drink of choice and set aside a few minutes to read it) article on music analysis and discovery software (last.fm, MusicIP, Pandora, etc…): Better Than We Know Ourselves
But in the past year or so– feeding off the iTunes-fueled growth of digital music sales– several companies have taken up the challenge, from Last.fm and Pandora, to the new wave of startups including MusicIP, Audiobaba, and Echo Nest. At present, these tools make rough and sometimes ludicrous recommendations; even at their best, they take some nudging and patience before they satisfy us. But they're getting smarter and brasher every month: They've grown from studying people's shopping carts to digging into the music itself and analyzing what makes it tick.
Music discovery tools could boost sales industry-wide– as much as tenfold, depending on who you talk to– and they're our best and, really, one of our only tools for tackling the marketing phenomenon known as the "Long Tail," where consumers wade through millions of niche and obscure albums thanks to the limitless shelves of online stores. Every day, bedroom musicians give away their Creative Commons-licensed mp3s, and digital distibutors snatch up forgotten back catalogs. San Francisco's IODA inked a deal to digitize 60,000 releases from China; how could any human wade through it all to find the best albums?
But how far can these tools go? Can they create maps of our sensibilities and tell us exactly what we want to hear– night and day– for any mood?
Recommended Links from Pitchfork:
- Last.fm: www.last.fm
- Pandora: www.pandora.com
- MusicIP: musicip.com
- Search Inside the Music: research.sun.com/projects/dashboard.php?id=153 (see also Paul Lamere: blogs.sun.com/plamere)
- Audiobaba: Audiobaba
- MusicStrands: www.musicstrands.com
- Echo Nest: www.echonest.com (see also Brian Whitman: variogr.am and Tristan Jehan: web.media.mit.edu/~tristan/)
- The International Conferences on Music Information Retrieval and Related Activities: www.ismir.net
I've been messing around with MusicIP for the past few hours, and it's really an amazing piece of software. It analyzes quickly, and it creates surprisingly consistent mixes. Pitchfork explains:
After the analysis, the mixer affixes a signature to the header of each file containing all the information it needs to compare that song with any other. The signature takes up a tiny 100 bytes– which is less memory than this sentence– and though the Mixer took hours to get to know your songs, when you ask it to come up with a playlist, it performs the task in a split-second. Start with Elvis Costello's "Clubland" and the Mixer will give you a playlist of bouncy pop songs. But start with atmospheric electronic music, and it'll match that as well. I wanted a mix that followed the hiccuping beat and eerie pauses of Boards of Canada's "The Devil Is in the Details", and the songs it pulled from my library– which ranged from David Sylvian to Kate Bush to Jorge Ben to Pulp– all fit that same strange tone, and often the same rhythmic pattern.
1 Comment