Tangerine!
Spent some time today playing with Tangerine!. Tangerine is an application that allows you to create playlists by analyzing the beats per minute (BPM) and beat intensity of the songs in your iTunes library. Once analyzed, you query the results by providing a desired range for both BPM and BI that you are looking for and voila…you have a playlist.
I have always been particularly intrigued by this as my musical tastes are rather eclectic and often specific to what I am doing at that moment. Which, of course, means that my playlists are difficult to create in any meaningful way as by nature they are static.
First thing I noticed was the speed in which it ripped through my library. Their blog states roughly 3 songs/second. on a 1.83 Ghz dual-core iMac. I am running this on my 1.67 Ghz PowerBook so did not see quite the performance but was able to process 3.2GB in 200 seconds which is roughly 2.5 songs/second. Frankly, this is faster than I anticipated.
I must say I have been pretty impressed with the consistency of the results. This is not to say that there are some oddities generated. For instance, one of the playlists consisted of Frank Sinatra’s Girl from Ipanema (BPM == 135, BI == 41) with Rob Zombie’s Dragula (BPM == 125, BI == 62). Also, some of the BPM’s that were reported by iTunes were in some cases significantly different from Tangerine!’s analysis. But being that this is in beta (more on that later) I am hopeful that the algorithm will continue to be refined.
A couple of other things you should note if you are thinking about giving it a try. It currently only runs on OS X 10.4 (Tiger). Also, due to the DRM encryption around songs purchased off of the iTunes store Tangerine! is not able to analyze them. Although, they have stated that they are actively working on a solution.
Finally, Tangerine! is currently in beta but if you blog about it or report a bug they will send you a license key when it is released. Here is the Potion Factory, developer’s of Tangerine!, blog.