Classical Music Gets a Search API

Peachnote is quite an innovative approach to music analysis. Developed by Vladimir Viro, a renaissance man who bridges the worlds of classical music and advanced mathematics, the Peachnote API provides the ability to search a large collection of public compositions by melody or sequence of chords.

This is a passion project run largely by one man who’s interest keeps it alive.  The site lists API functionality limited by access to sufficient hardware, but these limitations didn’t stop Viro from winning an award from Google in May 2011.

The documentation of the Peachnote API is very straightforward.  It’s all on one page.  Viro has taken a complex subject and provided value like any good API provider: by simplifying.  Doing something difficult or complex and making it accesible via API is a great strategy for relevance on the web.  There are three main search methods to this API:  Counting Ngram occurences, finding complete Scores, and pulling images of sheet music.  The count method allows you to tell how popular a particular melody has been through history.  By searching for scores,  you can access all the meta data for the piece that contains that melody.  Finally, if you actually want to see the music itself, you can pull up its digital image.

I’d be interested to hear from Vladimir on what his original intentions for the project were.  Is it a kind of due diligence for composers to check if the “new” melody they’ve created is actually new?  Is it the start of a classical music recommendation service?  I also wonder if the demand would support paid API access levels.  It’s always inspiring to see these projects and I hope this one continues.  It’s one of well over 100 music APIs in the index, but falls into a much smaller analysis category that will become more populated over time.

Be sure to read the next API article: Dropbox Woos Developers With New API Release


Comments (1)

Garrett, thank you very much for a nice review!

As to why I started developing Peachnote - I love music, and one of my quests in life has always been understanding how it works. Also I'm fascinated by data; for me it's like another element, like fire and water and air, I can watch it indefinitely. So trying to understand music by collecting and analyzing it was a natural thing for me to do. Sharing this joy with others is also quite nice - just ask the creators of Google Ngram Viewer. And of course there are plenty of interesting applications to be built on top of this data, I'm sure we will see many of them quite soon. I'll be attending the Music Hack Day in Boston this weekend, and hope that people will do something nice with the API. I don't know what, but that's the point of an API :-)

Hope to see you this weekend as well!