Uncategorized

Echo Nest Compares It’s Playlist Engine To iTunes Genius, Google Instant Mix

image from web.mac.com Listeners often prefer to have their music playlists created for them. Doing it yourself is just too time consuming. They also lead to new music discovery; so much effort has been put into machine generated playlists. Pandora has its version. iTunes Genius promised, but so far has failed, to revolutionize listening. And The Echo Nest created a music engine API that's powering dozens of apps from MTV to the BBC to We Are Hunted.  Last week, Google Music Beta jumped in with what TechCrunch called it's "killer feature" Instant Mix.  But how good are these data driven playlist creators and how do they compare?  

Paul Lamere, The Echo Nest's Director Of Developer Platform, did his own comparison, that could easily be called bias if it weren't so data driven and exhaustive. All  three are based on "secret" algorithms, but Lamere shared some basic difference:

  • iTunes Genius – "This system seems to be a collaborative filtering algorithm driven from purchase data acquired via the iTunes music store.  It may use play, skip and ratings to steer the playlisting engine.  More details about the system can be found in: Smarter than Genius? Human Evaluation of Music Recommender Systems.  This is a one button system – there are no user-accessible controls that affect the playlisting algorithm."
  • Google Instant Mix – "There is no data published on how this system works. It appears to be a hybrid system that uses collaborative filtering data along with acoustic similarity data.  Since Google Music does give attribution to Gracenote, there is a possibility that some of Gracenote’s data is used in generating playlists.  This is a one button system. There are no user-accessible controls that affect the playlisting algorithm."
  • The Echo Nest Playlist Engine – "This is a hybrid system that uses cultural, collaborative filtering data and acoustic data to build the playlist.  The cultural data is gleaned from a deep crawl of the web.  The playlisting engine takes into account artist popularity, familiarity, cultural similarity, and acoustic similarity along with a number of other attributes   There are a number of controls that can be set to control the playlists: variety, adventurousness, style, mood, energy."

See more of Paul Lamere's side by side comparison of these three top playlist engines on his Music Machinery blog.

Share on: