Announcing Echoprint
The Echo Nest has been focusing on a crucial component of the oncoming music cloud for some time: we spend a lot of time and engineering resources on music resolving. This extends from mapping a query for a band name to its ID, to uncovering mentions of songs on blogs, to identifying the song in an audio stream without any metadata - otherwise known as fingerprinting. The Echo Nest's existing fingerprint technology, 'The Echo Nest Musical Fingerprint' aka ENMFP, has been in wide use privately and via our API for 18 months. Today we are unveiling a new fingerprint technology called 'Echoprint,' whose main feature is its complete openness - everything from the program to analyze the audio to the server and data to make the match are available for anyone to use, under a permissive open source license, for free."
Posted Jun 23, 2011 16:08 UTC (Thu)
by lolando (guest, #7139)
[Link] (6 responses)
I nevertheless applaud the idea, and hope that this "omission" is only temporary and will cease to be in effect once some critical mass has been attained.
Posted Jun 23, 2011 17:51 UTC (Thu)
by cworth (subscriber, #27653)
[Link] (1 responses)
I assume you're referring to the license of the data they are sharing here.
As for the code, the code generator is being distributed under the MIT license and the server is being distributed under the Apache license.
So it looks to me like all the necessary code is available under DFSG-free licenses. And new services could be started with various licenses for the data itself.
-Carl
Posted Jun 23, 2011 19:17 UTC (Thu)
by lolando (guest, #7139)
[Link]
Yes, my bad, I should have been more precise. Hower, one could argue that the license of the actual code may be less relevant than the one for the dataset. If the algorithm is public, then the code can be reimplemented in a compatible and interoperable way. But if we have several datasets under incompatible licenses, there's a networking effect if only one dataset can be under use at any one time. There could, however, be an alternative, clean-room dataset under a license that doesn't fail the "desert island" criterion; if the clients are made to query several data sources, then the initial dataset becomes less important indeed (as long as there's no leakage of data).
Posted Jun 30, 2011 14:01 UTC (Thu)
by stevem (subscriber, #1512)
[Link] (3 responses)
Posted Jul 5, 2011 15:17 UTC (Tue)
by nye (subscriber, #51576)
[Link] (2 responses)
I've never really thought about it before, but it sounds on its face like a good idea. What are the objections to it?
Posted Jul 5, 2011 15:35 UTC (Tue)
by mjg59 (subscriber, #23239)
[Link]
Posted Jul 5, 2011 15:54 UTC (Tue)
by stevem (subscriber, #1512)
[Link]
In my opinion all three of these tests are broken precisely because they're trying to apply hard tests to a license. This is explicitly not in the spirit of the DFSG.
Posted Jun 23, 2011 16:09 UTC (Thu)
by clint (subscriber, #7076)
[Link] (1 responses)
Posted Jun 25, 2011 15:43 UTC (Sat)
by dr@jones.dk (subscriber, #7907)
[Link]
Echoprint data makes it possible to auto-generate a playlist. Chormaprint makes it possible to identify a tune even if degraded by lossy compression.
Announcing Echoprint
Announcing Echoprint
Announcing Echoprint
"Desert Island" test
"Desert Island" test
"Desert Island" test
"Desert Island" test
How does it compare to acoustid/chromaprint?
Announcing Echoprint
Announcing Echoprint