Why study Indian music?
I am often asked why I tend to focus my application on Indian music. This first reason is that Indian music encompasses a vast array of important musical styles. Second, because the underlying melodic and rhythmic frameworks are common to a great deal of music from South Asia, the Middle East and North Africa. Third, I believe that advances in music technologies will be sparked by careful examination of specific musical traditions. For example, we cannot hope to model improvisation generally without working carefully to understand specific improvisational traditions. In my opinion, general attempts either reveal an implicit bias on the part of the researcher for a particular musical tradition, or are hopelessly general and without real content.
Perhaps more fundamentally I believe that most machine learning algorithms will have to learn to deeply understand and represent musical context in order to be successful. By working at this problem from a variety of angles we may better be able to see the general form of context dependency. My intuition is that pursuing it in specific domains and synthesizing these insights will prove to be a more successful strategy than attempting at the outset a fully general model. For example, in recommendation work, which is fundamentally about similarity, we are attempting to exploit the particular characteristics of Indian music that make certain types of analysis possible. For example, it is easier to build melodic models, in addition to the usual timbral models, because Indian music tends to be much less polyphonic than Western music.
Here are a some examples of music that I analyze and that share, despite many surface differences, fundamental melodic and rhythmic concepts.