song/identify supported both ENMFP and Echoprint, and AFAIK the Echoprint matching path was exactly the same as is published on Github.
I know at some point we did adapt the Solr end (for example, we removed the N most occurring codes) for speed optimizations.
Many users of Echoprint in the wild have adapted the python matching logic for their use case as well as changed the hash update rate on the codegen. A great modification to watch was Sunlight Labs' "Ad Hawk", which ID'd commercials: https://github.com/sunlightlabs/adhawk
I like this concept, but you need some sort of automated filtering. My first track (because hypem liked it) is a cat purring.
Plug but it's a useful one: run the tracks through the SCAnalyzer http://labs.echonest.com/SCAnalyzer/index.html and filter for speechiness and duration. If you get a SO ID, you'll also know the artist via fingerprinting and you can do more filtering at that level.
we're adding j-core now, thanks for the suggestion!
sxsw is why this is special, we will be adding and changing these over time. It's clear that genres don't have to directly correlate with musical style.
These genres come from people talking about music using whatever organic / natural words they use to describe music. We've done the work to map those into single-term "genres," as it's been shown many listeners appreciate the flat categorization. There are similarities between genres in the API, so you can quickly see which genres sound like others.
Analyze your own, it'll be much better when you're doing synchronous stuff like this. Our canonical audio is likely very different in bitrate or time to start.
Yes: you can use Echo Nest Remix in python or do what paul did and pre-load the analysis and do the playback via webaudio. Remix takes care of the chopping at the right bits for you: http://echonest.github.com/remix/
I know at some point we did adapt the Solr end (for example, we removed the N most occurring codes) for speed optimizations.
Many users of Echoprint in the wild have adapted the python matching logic for their use case as well as changed the hash update rate on the codegen. A great modification to watch was Sunlight Labs' "Ad Hawk", which ID'd commercials: https://github.com/sunlightlabs/adhawk