Sander Dieleman, a PhD student at the University of Ghent in Belgium, is working on music discovery algorithms and is also interning at Spotify in New York – and now he has blogged about what he is doing there and why. In order to unearth for listeners both new and old tracks they might like, he is focusing on their sound rather than following the old “wisdom of crowds” model that dominates music discovery (i.e. pushing a listener towards the music that others with similar taste are listening to). Hold up – so essentially it’s treading the same path to Pandora’s Music Genome project that has been running for over a decade? Well, kind of. Dieleman has analysed 30-second clips of half a million of the most popular tracks on Spotify to understand their acoustic characteristics and serve recommendations around that. Music streaming services are betting the farm on discovery. Spotify bought The Echo Nest earlier this year, Rdio acquired TastemakerX soon after, Apple snapped up Beats Music and just this week Rhapsody acquired both Ex.fm and SoundTracking. They are all craving that essential point of differentiation as, beyond the odd exclusive, they have the exact same catalogues. So discovery and recommendation are where the USP arms race is happening. For now.
Tools: platforms to help you reach new audiences
Tools :: We Are Giant
With “fan communities” being on every artist’s team’s mind, we’re fans of the fact that…