Day #2 at the startup

I just finished day #2 at my new job.  Sorry to be so cagey about where I was going, but they wanted to keep it quiet until they could  do a press release about it.   I see the press release is public now, so I’m free to talk about my new job.

As many of the commenters have guessed, I’ve joined, as  the director of the developer community,  The Echo Nest -   a company that is devoted to providing music intelligence for the next generation of online music application.   In this role, I will work with the rest of the  Echo Nest team to help grow an active, vibrant music application developer  community around The Echo Nest  developer API.

I’m really excited to be here at The Echo Nest.   The Echo Nest has already established a reputation as a company that provides a new breed  of hardcore music intelligence.  The Echo Nest goes far beyond the “wisdom of the crowds” model of music discovery (“People who listened to the Beatles also listened to the Rolling Stones”). Instead of just data mining user behavior, The Echo Nest crawls the web to learn everything it can about music by analyzing what the world is saying about music.  The Echo Nest also directly analyzes the audio content of music – extracting musical traits such as key, tempo, structure, timbre from the audio.

Home of The Echo Nest

Home of The Echo Nest

From this  analysis of the social context, the user behavior and the actual audio, the Echo Nest gets a deep understanding of the entire world music.   It knows which artists are getting the most buzz, which artists are getting stale,  how and why artists are related, what words are being used to describe the music.     This data goes far beyond the “if you like Britney, you might like Christina” level.  The Echo Nest understands enough about music to be able to answer queries such as “make me a playlist of  songs with a tempo of 90 beats per minute by an unknown emo artist that sounds something like Dashboard Confessional,  and has violins”.  The really neat thing is that the Echo Nest is exposing a lot of this functionality in their developer API.  This lets anyone who is building a music application to tap into this large resource of music intelligence.

One of my main duties is to be the voice of the developer in the Echo Nest.  I’ve written my fair share of music apps, so I have a good idea of some of the many pain points and difficulties that a music application developer has to face, but I’d like to hear more, so if you are developing a music application and you need a particular problem solved let me know – or better yet, post to The Echo Nest developer forums.

I’ll be writing a lot about The Echo Nest in upcoming posts – in particular about using the developer APIs, but I shall still continue to post about all of the interesting things going on the music space – so this blog won’t be too much different from Duke Listens!

  1. #1 by Zac on February 20, 2009 - 8:46 am

    Hey Paul, nice work.

    I did a mini-review of the EchoNest-powered ThisIsMyJam (jeez) almost a year ago, as well as sharing a couple e-mails with Adam Baratz.

    http://datawhat.blogspot.com/2008/03/share-your-musical-taste.html

    Impressive even then. can’t wait to see what’s next.

  2. #2 by Luke Matkins on February 20, 2009 - 1:52 pm

    Congratulations and best of luck to you guys!

  3. #3 by Josh on February 20, 2009 - 10:35 pm

    Sounds awesome! I’ve been hoping someone would develop a project like this for years – after finding just about every music recommendation engine to function like you mentioned – “if you like Britney, you might like Christina” which isn’t that great.

    See what I don’t understand is how someone like me can listen to almost no dance music – yet on rare occasions I will hear a dance song I like. If a computer could identify exactly what it is about that song I liked, when in general I don’t like that kind of music than that would be really impressive.

    Is that the kind of thing The Echo Nest will be able to do?

  4. #4 by jeremy on February 23, 2009 - 2:05 pm

    Mmm.. content-based music similarity measures! My favorite!

    Seriously, this is really good that Echo Nest is rolling all this out.

  5. #5 by Dan Foley on February 24, 2009 - 12:08 pm

    I was somewhat unhappy when I saw that Duke Listens was ending, but I’m glad to see that the story continues!

    I would actually like to hear that playlist with the violins. I remember being quite interested in the echo nest when I first heard about it, so I’ll be keeping an eye on developments here too…good luck with the new job!

    http://www.podcomplex.com/blog/music-technology-update-echo-nest-hatches/

  1. Music and Exploratory Search « Information Retrieval Gupf
Follow

Get every new post delivered to your Inbox.

Join 1,094 other followers

%d bloggers like this: