Mood predictor for iTunes

Smart Playlists in iTunes are incredibly useful, but require quite a bit of time and thought. Usually when I’m in the mood to hear a bunch of songs by a select few artists, I’ll manually pick out songs by those artists and drag them into Party Shuffle because the process of creating Smart Playlists simply takes too long.

Mood, a new “special playlist” in the same vein as Party Shuffle, would do all the Smart Playlist building for you. All you have to do is give it some clues by dragging in songs you definitely want to hear.

You start by dragging some handpicked songs into the Mood playlist in the Source menu on the left side of the iTunes window. The songs you pick will define your mood, and the Mood algorithm would be able to predict similar songs you might want to hear based on meta criteria that the iTunes application already logs.

Information like release year, genre, skip count and play count are already logged by the application. The Mood playlist would look for similar characteristics of the songs you choose and deliver songs that match a logically similar criteria.

For example, let’s say you start by dragging a couple Flock of Seagulls songs. The Mood playlist begins by filling with more songs by that artist. Then you add a Corey Hart single and some tracks from Michael Jackson’s “Off the Wall.” The mood generator will see that they’re all 1980’s popular songs by comparing all of their meta tags, seeing the similarities in year and genre. It would then populate the list with more songs that match the 80’s pop song criteria.

To change the mood, click on “New Mood” at the bottom of the Mood playlist. The list is empty, and you can begin dragging a new set of songs into the Mood playlist.

Using play count criteria, it might notice you choose all songs you haven’t heard before and will find others. Let’s say whenever you’re feeling depressed, you play a lot of Elliott Smith and Pinback. When you start adding songs by those artists to the Mood playlist, a message will popup saying “get laid” it will notice that all of them were played around the same time a few months ago, using the Last Played count and suggest other songs or similar ones that were listened to during that time.

As mentioned in the beginning of the post, all of this functionality could be accomplished by the enduser simply by noticing the trends in the songs he is choosing and plugging that info into a Smart Playlist. But this is a more user-friendly way doing all the analysis, making the user think as little as possible.

Advertisements

2 comments so far

  1. tunesfan on

    Check out MoodShuffle at http://www.moodshuffle.com It does pretty much what you are looking for, but in real time as you run it. It doesn’t mess with playlists, although the website says the ability to save them is coming in a future version. It has a real simple interface. When you want to change your mood, you just quit, and restart the app. Pretty slick.

  2. […] year ago I posted about the idea for an iTunes “mood predictor” that watches your listening habits and adapts the playlist to your current mood. It seems there was […]


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: