music information retrieval (mir)

I’ve been interested in learning more about music information retrieval systems. here are a few resources i’ve been looking at:

*There’s a Google Tech Talk called “from sound synthesis to sound retrieval and back” by Xavier Serra of Music Technology Group at the University of Pompeu Fabra in Barcelona. Serra has quite a few projects in development from sound retrieval systems that look for similar harmonic structure or rhythm to developing tangible interfaces. He also demonstrates the “reacTable” which is an instrument that Bjork used on her Volta tour.

Some really cool things happening with MIR. With signal processing, users can access more than just what libraries normally catalogue (info attached to the physical format, digital info like text, digital sound data) and go further up the symantic ladder to obtain info like pitch, duration, timbre, intensity and further up the ladder as high as type of music, key of music, melody within a polyphonic piece, etc.

*“Information Retrieval for Music and Motion” by Meinard Muller is one of those good Springer books. I’m just getting into it now, but it looks pretty comprehensive. Two major parts: part 1 Analysis and retrieval techniques for music data; and part 2 Analysis and retrieval techniques for motion data.

Advertisements

4 thoughts on “music information retrieval (mir)

  1. Hey Jan,

    Cool stuff – I hadn’t seen that book before. I’ll have to give it a read. (I notice McGill’s copy is checked out of the library… hmmm…. 🙂

    If you’re hardcore about it, you can also peruse all of the ISMIR proceedings online. There’s tons of stuff there, on all aspects of MIR:

    http://tinyurl.com/55x4qo

    -A

  2. jandawson says:

    Thanks Karin, Midomi is great! Been playing with it tonight. It returned the right answer as the first result 9/10 times so far. (I had to give it something hard so i tried bohemian rhapsody and it returned i’m too sexy by right said fred 8) hahaha.)

  3. The Melody Catcher is an accurate melody finder because its matching technique is
    focused on monophonic and well defined on-line audio sources, found by a Web crawler. In the results the user is linked back to the source of the wanted melody

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