April 3, 2008 In a quest to find "the absolute least amount of data needed to reproduce a piece of music", researchers at the University of Rochester have digitally encoded a 20-second clarinet solo into a file that's less than a kilobyte in size - nearly 1,000 times smaller than standard MP3 compression.

But it's not about cramming more onto your iPod. The research funded by the National Science Foundation aims to give computer musicians more intuitive tools with which to create expressive music - a goal that ties in with the method used to achieve the ultra-compressed file. This method involved duplicating the physics of both a clarinet and a clarinet player in a computer and then teaching it to "play" using a recording of a real performance. This virtual musician method not only delivers the tiny file that according to researchers is "close, though not yet a perfect", but also means that computer synthesizers can be made to incorporate the actions of a real-world musician.


Upgrade to a Plus subscription today, and read the site without ads.

It's just US$19 a year.


"This is essentially a human-scale system of reproducing music," says Mark Bocko, professor of electrical and computer engineering and co-creator of the technology. "Humans can manipulate their tongue, breath, and fingers only so fast, so in theory we shouldn't really have to measure the music many thousands of times a second like we do on a CD. As a result, I think we may have found the absolute least amount of data needed to reproduce a piece of music."

Read more and listen to the result at the University of Rochester site.