Software that ‘listens’ to songs could help you find new favorites
Sure, you’ve heard U2. Hard to avoid the Irish rockers, a musical force since the 1980s.But how do you find – let alone decide if you like – some garage band that’s cut the killer new track that no one’s ever heard of?
In his gleaming white lab at Drexel University, Youngmoo Kim has an answer: raw computing power.
He’s an electro-DJ of sorts, part of a new wave that seeks to help consumers sift through the countless downloadable tunes on the Internet. It’s a crowded field, with plenty of Web sites already promising consumers “if you like this, then you’ll like that.”
But most of those sites base their picks largely on what other consumers have purchased – perhaps reinforcing mass-market tastes at the expense of undiscovered gems. Kim and other researchers have embraced a more fundamental approach: using computers to “listen” to the music itself.
:: 
:: 
:: 
:: 
:: 
:: 
:: 
:: 
:: 
:: 
::Tullycast:: 