Machine Learning and MP3s

dan moore writes "Students at Caltech and Harvard have developed a system that analyzes playlists and learns people's listening patterns. It then channels its knowledge into generating streams of music that the people themselves would like to listen to. Intuitive, accurate, and finally someone has done it. Check out the website to get one of the available plugins. Another interesting approach to digital music."

Others...

[moeffju]

For other programs that do this already, look for RoboDJ or AudioScrobbler. Lots of others exist.

Yet none get the job done right.

Re:But people are lazy...

[chickens]

"find out what other people who really like this song listen to" programs..
Methinks you'd like audioscrobbler [audioscrobbler.com], which is somewhat like firefly

Re:Random playing

[iksowrak]

I just found out about the plugin from /. this morning and haven't read up much on it yet, but it appears to factor in how long each song is played. So if you're like me and have Winamp on random play but skip over (or partway through) songs I don't feel like listening to, the plugin will still do its work.

Already Exists!

[captainclever]

There is already something like this out there, that uses loads of cool independant and smaller label stuff as well as some massive label stuff..

Check out Last.FM [last.fm], they are very good. I've found a load of new artists from there. It is all stream based (128kbps) and they have a massive flash development section starting for open source goodness.

There's also the (all open source) Audioscrobbler project.. see my .sig :)

RJ

Smart Playlists?

[ihatewinXP]

Another 'cutting edge playlist technology" would certainly be iTunes 'smart playlists.' You can match any number of criteria, including: genre, my rating, play count, artist contains *, and year to make sick and incredibly easy playlists. Oh and live updating, perfect for running a PlayCount: Zero and then having it add new unplayed tracks as you listen. At first I didnt notice it but after tinkering around I now wonder what I ever did before (but then again I get that feeling alot using apple products).

Check it out: http://www.apple.com/itunes/smartplaylists.html

Anyone heard of Last.FM

[Anonymous Coward]

Last.FM [last.fm] is great, it starts streaming you music, you skip the songs you dont like, it learns what you like, and your personalised stream gets better and better!

Check 'em out for some great new music from independants and small/medium label stuff :)

Re:Others...

[foxcub]

I geuss you could add GJay [sf.net] to the list...

Re:Machine Learning parts

[AssFace]

I have no clue what their software does "behind the scenes" - but I personally would use a Markov Matrix/Chain and Bayesian classification.

I haven't used their system - but if it generally sucks if not many people have used it, and then gets much "smarter" over time and as more users (and usage) increases - then I would suspect that is what they are using as well.

Essentially you have song A, and then that points to a list of songs (after listening to song A, people then tended to listen to song Z, Y, and F).
Songs that are more frequently occurring in that list are ranked higher.
You could then follow that (basically just a hash) and say that if you listen to song A, the most frequently listened to song after A is Z, then you go to the Z spot in the hash and see what it is most frequently followed by.

Re:I don't WANT to hear the same crap over and ove

[captainclever]

last.fm [last.fm] might be for you then.

Re:Listeng tastes

[bornholtz]

Most songs that make it onto my playlists are because a close friend recommends it

That's one of the reasons that I use AudioScrobbler [audioscrobbler.com].
My brother lives about 500 miles away from me and we can see what each other is listening to. I'm pretty comfortable listening to just about anything in his playlist.
He's a freshman in college and I'm an old fart. This allows me to learn about a lot of new music.

Re:this guy is one to watch...

[Hal-9001]

Dunno, but I've heard anecdotally (and the anecdotes were supported by a quick Google search) that he's on Caltech's ACM programming team, so he's definitely one smart cookie.

Also, he has a Slashdot account: SkyIce [slashdot.org]

Re:Others...

[cgroom]

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.

Some clarification (from one of the programmers)

[Mark Zuckerberg]

I made the program with Adam D'Angelo, which you can verify on the team page [synapseai.com].

I want to offer the following information about the project. The majority of development effort went into building the Brain. The Synapse player is just something we threw together to get the most from the Brain's functionality. We will probably never port Synapse to other systems since more than enough players already exist. Synapse does work under Wine though. We do have plans to bring the Brain to other systems, and we've begun by writing a plugin for Winamp 2.x, which you can get here [synapseai.com]. An XMMS plugin is coming soon, and then hopefully there will be one for iTunes in the near future.

And a note about privacy. None of your musical listening data will be available to anyone other than you. We hope to use massive amounts of data to aid in analysis, but your individual data will never be seen by anyone else.

Re:this guy is one to watch...

[the great chan]

It looks like they're both college freshmen now. But last year, Adam D'Angelo went to Korea for the IOI contest. Apparently, the other one is a smart guy too. A friend at Exeter said Mark Zuckerberg was a bigshot in math there and had some interesting coding projects of his own. Go figure.

So far I'm not impressed...

[jerkychew]

I downloaded and installed Synapse. Damn thing scanned all my local and network shares for files, giving me no option to manually specify. Since I have lots of drives, this took a good ten minutes.

What's worse, the app stayed on top of all my other apps, smack in the middle of my primary monitor, with no way to move it. So, I had to work on my secondary while it chugged away.

Ok, fine. Told it where the MP3s were, and it imported all the song info. I believe I was allowed to move this window, although I can't totally remember.

Fired it up. Black on dark blue background. Um, tough to read to say the least. Switched to the 'playlist' screen and tried dragging n dropping an m3u playlist into the screen, a la winamp. No dice, wouldn't load.

Ok, can't find any place to manually add files without exploring the little music database it built. Open the database tree and, holy crap, what a piece of shit. I wanted to listen to Linkin Park's Meteora CD, so I scroll waay down to linkin park, and expand the tree. Ugh. Flat file listings, by song name. Crap. Can't find Meteora.

Now, I know that this is kind of a different MP3 player, and I had every intention of RTFMing before really using it, but come on. It should at least be intuitive enough for me to be able to load some songs without having to read the instructions.

I closed Synamps and fired Winamp 3 up. Maybe I'll go back and try it out again, but I'm not as interested as I was when I started.