<body><script type="text/javascript"> function setAttributeOnload(object, attribute, val) { if(window.addEventListener) { window.addEventListener('load', function(){ object[attribute] = val; }, false); } else { window.attachEvent('onload', function(){ object[attribute] = val; }); } } </script> <div id="navbar-iframe-container"></div> <script type="text/javascript" src="https://apis.google.com/js/plusone.js"></script> <script type="text/javascript"> gapi.load("gapi.iframes:gapi.iframes.style.bubble", function() { if (gapi.iframes && gapi.iframes.getContext) { gapi.iframes.getContext().openChild({ url: 'https://www.blogger.com/navbar.g?targetBlogID\x3d9878398\x26blogName\x3dTom+Conrad\x26publishMode\x3dPUBLISH_MODE_BLOGSPOT\x26navbarType\x3dBLUE\x26layoutType\x3dCLASSIC\x26searchRoot\x3dhttps://tomconrad.blogspot.com/search\x26blogLocale\x3den_US\x26v\x3d2\x26homepageUrl\x3dhttp://tomconrad.blogspot.com/\x26vt\x3d6886542566053880231', where: document.getElementById("navbar-iframe-container"), id: "navbar-iframe" }); } }); </script>

Tom Conrad

I've moved my blog to http://tomconrad.net. Visit me there.

Wednesday, January 19, 2005

Savage Beast: Music Recommendation

Since Thomas Hawk mentioned this blog and my company [Savage Beast] in a recent post, I thought I'd tell you a bit about what we're doing here.

Savage Beast got its start 5 years ago with a mission to create software to help people discover music they will love. Have you ever spent enough time in a record store that the personnel started to develop a sense of your taste and started to make recommendations? Maybe you've even been lucky enough to have someone stash stuff behind the counter for the next time you visit the store? At Savage Beast, we're working on systems that will give users that same experience, but in the digital domain. We're particularly passionate about helping you discover artists out on the "long tail" (blog).

How do we do this? We believe that your musical preferences are first and foremost derived from the way music sounds. So, to help you find music that sounds like the music you like, we've spent the last 5 years working on something we call the Music Genome Project. The Music Genome is a detailed analysis of a century of popular music. We employee a large team of professional musicians who spend their days listening to music one song at a time and "analyzing" the detailed musical characteristics of the song. This includes obvious stuff like tempo and key, but more importantly it covers a lot of subtle nuance: how much tremolo in the voice, what kid of harmony, guitar picking style, how much cowbell, etc. In total we listen for hundreds of different qualities and painstakingly record the results for each some. After 5 years, we've built up an incredible asset that allows us predict music you will like based on simple input (for example, the name of an album, artist, or song that you love).

While we've been building this database we've been working with various partners and customers to deliver recommendations and music discovery experiences to end users. For example, we've built a complete retail kiosk solution for Music Discovery that have been deployed for pilot by Best Buy and Borders. AOL is also a customer and uses Savage Beast technology to power recommendations within their music properties. You'll see us branching out into new opportunities as 2005 goes on.

If you have a passion for music and are an exceptional software engineer, I'd love to talk to you about a job on our engineering team. Feel free to write to me directly at tomconrad-at-gmail.com.

Please note: this blog has moved to http://tomconrad.net, please do not post comments here.

At 5:25 PM, Blogger Marc Novakowski said...

Definitely needs more cowbell.

At 11:53 AM, Blogger Unknown said...

I really want you to explore the space...

At 8:31 PM, Blogger gfm said...

Related: GenieLab


Post a Comment

<< Home