I've been thinking ways of doing playlist trajectories. It's a tricky problem, though, because the obvious solution is to use the BPM, but that's not a great indicactor of chillness. I think the best way would be to morph between two user-defined "colors" (e.g. start light blue, move to dark red). But this would require (minimal) human intervention. Still, it's quite interesting.
GJay had scaled well on my system (50Gb collection). The only slowness comes at load time when it parses the 5Mb XML file. *Shrug*
GJay is a playlist generating app that pretty much does this. It forms playlists based on a least-squares fit on songs' frequency "fingerprint" (FFT), BPM, and user-assigned "color" (not required). This suffices to automatically generate interesting playlists across large music collections. Of course, you have to pre-process your music collection, which can be done in a background process. There is no AI, just some intelligent matching on both song characteristics and user categorization.
Surprisingly, matching on the frequency distribution between two songs gives a decent match between songs. A pure fingerprint match aligns songs which use the same spectrum range, which generally translates to similiar mood and instruments. This is of course not enough to go on, which is why GJay will also factor in the BPM, user 'color' rating (if supplied), and how you filed the songs in the first place. The spooky thing is that GJay tends to do a very good job at generating playlists with little prior knowledge of your listening habits.
(Caveat: as the author, I'm biased).
I'm sorry I haven't gotten around to a Windows version. If there's interest/help, it's definately a possibility.
Slashdot Mirror
GJay had scaled well on my system (50Gb collection). The only slowness comes at load time when it parses the 5Mb XML file. *Shrug*
GJay is a playlist generating app that pretty much does this. It forms playlists based on a least-squares fit on songs' frequency "fingerprint" (FFT), BPM, and user-assigned "color" (not required). This suffices to automatically generate interesting playlists across large music collections. Of course, you have to pre-process your music collection, which can be done in a background process. There is no AI, just some intelligent matching on both song characteristics and user categorization.
Surprisingly, matching on the frequency distribution between two songs gives a decent match between songs. A pure fingerprint match aligns songs which use the same spectrum range, which generally translates to similiar mood and instruments. This is of course not enough to go on, which is why GJay will also factor in the BPM, user 'color' rating (if supplied), and how you filed the songs in the first place. The spooky thing is that GJay tends to do a very good job at generating playlists with little prior knowledge of your listening habits. (Caveat: as the author, I'm biased). I'm sorry I haven't gotten around to a Windows version. If there's interest/help, it's definately a possibility.