The Echo Nest's company blog on music data, fingerprinting, genre analysis, and its API, ending with its acquisition by Spotify.
HN links it for the Spotify acquisition and genre classification; commenters recall Music Hackday and note the API would stay up.
Oh, they have been aware of it for a long time.
https://blog.echonest.com/post/48943428838/how-we-cope-with-spammers-fakers-and-cloners
Spotify is actually half-decent here. The other streaming services seem to mainly compete on cutting costs, and many are little more than frontends over 7digital's services (who, surprise, also do virtually nothing to prevent spam).
To take an example:
https://www.deezer.com/search/%22From%20the%20box%22/album
That particular spammer has been doing this thing for years. He's easily recognizable: one characteristic is the way he uses colour filters. If you search qobuz (one of these terrible front-end streaming services which does no QQ) for the label "piano to go", and say, the album "Joker Games", you'll notice it's uploaded once for Bill Haley, and once for Bill Haley & his Comets. That was apparently close enough to be detected as a duplicate, so he put a blue image filter on the latter.
Most of this guy's albums eventually get deleted (or delisted), even on sites like Qobuz - probably more for lack of plays than for being detected as spam, I'd bet. So you don't get his full history. But I found out the French library service made some effort to catalogue streaming releases a while back, and with access to search in those, you can track his evolution. He was a lot sloppier in the past, for instance he didn't bother to come up with a new label name for every dump. But he's been at it since at least 2015.
Really happy for these guys. :) I consulted with Norwest for technical due diligence during the last fundraising round[1], and was super impressed with what this team has built. Audio classification at scale is a really difficult thing to get right.
They're also one of the few companies who have successfully navigated the minefield of the music business, which takes a painstaking amount of focus and determination. But above all, they created a remarkable service to powering the next-generation of music experiences. Spotify seems like an obvious fit (especially given the Spotify Platform[2]), and I'm really excited to see where they go next.
As an aside, The Echo Nest has been a huge supporter of the start/tech scene in Boston. For 5 years now they organized "Music Hack Day"[3] which brought together musicians and engineers to build fun projects. Here's hoping they keep it up!
[1] http://www.forbes.com/sites/tomiogeron/2012/07/12/echo-nest-raises-17-million-for-big-data-analysis-of-music/
[2] https://developer.spotify.com/
[3] http://blog.echonest.com/post/66776489063/the-echo-nests-5th-music-hack-day-boston-was-a-roaring
1. I can name exactly one time that an ad was actually something I might be interested in.
2. Conflict of interest. An advertiser's job is to get me to buy/view/etc something, often with misleading statements. Whether me doing that is actually in my best interest isn't factored into the equation.
3. Annoying. 'Shady' ad practices do and will continue to exist. In any case they take up screen real estate.
4. Advertisers are compiling huge databases that they intend to fill with all sorts of intimate information about me. What I like to read (what I read), who I talk with, what I talk about, etc. These databases are ripe for all sorts of abuse, both outside and within the domain they're being compiled for. (Great example: http://blog.echonest.com/post/27047918145/musical-taste-politics)
5. Another conflict of interest is between me and the service itself. When the service is mainly powered by ad revenue it means that anything that's good for ads is in the interests of the service, even if it's bad for me. (Eg. Taking up more and more screen real estate, sandwiching ads in between the 'legitimate' content.)
For a start.
The Echo Nest - Somerville, MA
The Echo Nest powers amazing music apps with our API, including apps from the BBC, MTV, iHeartRadio and other music services you use and love.
We've got a killer engineering culture, well-stocked beer fridge and you can help shape the future of music. Come meet the whole company and see what it's like this weekend at Boston Music Hack Day: http://boston.musichackday.org/2011/
The after party is at our offices and Questlove is playing. How many of your company parties has Questlove played?
Jobs:
Engineer
Junior Engineering Intern
Music Information Retrieval Scientist
Senior Engineer
http://the.echonest.com/company/jobs/
Music Hackday was great (Infinite Gagnam Style anyone? [0]) and I especially remember Reykjavik 2012 as a personal transformative event! I made Horisfy [1] and and won best hack, very similar to Johnny Cash, but riding horses on Iceland.
This was the time before Echonest was bought by Spotify and Tomahawk (used in Johnny Cash hack) was sprawling.
Fun memories :=)
[0] https://blog.echonest.com/post/34826507401/infinite-gangnam-style
[1] https://web.archive.org/web/20121031194040/https://wiki.musichackday.org/index.php?title=Horsify
See also http://everynoise.com/ which is a view into how Spotify classifies music.
The creator wrote about it here:
http://blog.echonest.com/post/52385283599/how-we-understand-music-genres
and writes a lot about it on their blog:
http://www.furia.com/page.cgi?terms=noise&type=search
Of course those are going in the other direction, not generating the classification from the data, but it's probably one of the best data sets as far as classifying existing music.