Exploring the Echo Nest API

I’ve spent all day today geeking out on APIs. Still mucking around with The Echo Nest API, and testing things out bit by bit. I’ve got a handle on pretty much how to go about getting data and such… but some of the particulars are eluding me.

While I understand the methods and mostly understand the data they return, I’m still unclear as to HOW one goes about uploading a song to be analyzed. Do I uplaod to Echo Nest directly? If so where? And how can I access that uploaded file for playback?

In digging around a bit more, looking under the hood at both The Jingler and More Cowbell, I think I’ve figured out roughly how the process works:

1) The sites allow users to upload their own mp3s.
2) The mp3 is uploaded over to Amazon Simple Storage Service (aka Amazon S3)
3) The file is referenced based on URL, and subsequently analyzed.
4) The file (on Amazon) is accessed for playback, while data is accessed from Echo Nest.

I’m currently futzing around with the getSegments method. And while I’m able to pull down data… I’m a little unclear what to do with it all. Timbre? Pitch? I found what I think is an older/outdated page, but the descriptions are still eluding me.

I studied music as a kid (violin, piano, guitar, choir), and know my way around music stuff fairly well. Better than your average joe, at the least. But things like tatums and 12-number chroma vectors are outside my understanding.

There’s all this STUFF there, with the Echo Nest API. And I’ve got access to it. I just don’t really quite grok exactly what I have access to, you know?

Guess I’ll just keep checking in on their forums. It’s a bit of a ghost town, but it’s all I’ve got at the moment.

Still wishing I could take a week off of work to play.