am I using the "extension" field correctly?

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

am I using the "extension" field correctly?

Nathaniel Beaver
I've extended XSPF to include Chromaprint fingerprints as described here:

https://github.com/nbeaver/chromaprint-playlists

Am I using the extension field correctly?

Nathaniel Beaver
_______________________________________________
Playlist mailing list
[hidden email]
http://lists.xiph.org/mailman/listinfo/playlist
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: am I using the "extension" field correctly?

Lucas Gonze-2
Yes, the syntax of your usage is correct.

This is a really good idea for an extension. Nobody has done a fingerprint yet.

About the idea that there is no fully functional content resolver
available yet, I don't think of this as a binary yes/no. Resolvers are
analogous to search engines which may return only a single result.
Over time the results have slowly gotten better as one project
improves on the state of the art. Your project has the potential to
push the state of the art once again. Pretty awesome.

The best current resolver AFAIK is Tomahawk. Those guys would totally
appreciate what you're doing.

And obviously the Musicbrainz faction would also get it.

What will be difficult for you is developing a large catalog of
fingerprints. Have you checked in with the MB folks? They used to do
fingerprinting. And it would be natural to use MB IDs for the
underlying catalog.

Happy to make introductions.

On Tue, Dec 22, 2015 at 4:00 PM, Nathaniel Beaver
<[hidden email]> wrote:

> I've extended XSPF to include Chromaprint fingerprints as described here:
>
> https://github.com/nbeaver/chromaprint-playlists
>
> Am I using the extension field correctly?
>
> Nathaniel Beaver
> _______________________________________________
> Playlist mailing list
> [hidden email]
> http://lists.xiph.org/mailman/listinfo/playlist
_______________________________________________
Playlist mailing list
[hidden email]
http://lists.xiph.org/mailman/listinfo/playlist
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: am I using the "extension" field correctly?

Nathaniel Beaver
On 12/23/2015 01:04 PM, Lucas Gonze wrote:
> Yes, the syntax of your usage is correct.
>
> This is a really good idea for an extension. Nobody has done a fingerprint yet.

Thanks! It was pretty straightforward because most of the work had
already been done.

>
> About the idea that there is no fully functional content resolver
> available yet, I don't think of this as a binary yes/no. Resolvers are
> analogous to search engines which may return only a single result.
> Over time the results have slowly gotten better as one project
> improves on the state of the art. Your project has the potential to
> push the state of the art once again. Pretty awesome.
>
> The best current resolver AFAIK is Tomahawk. Those guys would totally
> appreciate what you're doing.

I'd heard about Tomahawk before, but haven't actually installed it until
you suggested it. I must say I'm impressed. The ability to transparently
pull from multiple sources for a single song is something I've been
wanting in an audio player for a long time.

One area where the acoustic fingerprints could help would be in
preventing censorship (or enforcing it, I suppose). For example,
Kendrick Lamar's "King Kunta" is listed on Spotify twice, once with
censored version:

https://open.spotify.com/track/043Rs5H7XAV2soPAy88x02

and again with the "explicit" version:

https://open.spotify.com/track/0N3W5peJUQtI4eyR6GJT5O

Unfortunately, the album name is the same ("To Pimp A Butterfly") and
the track name and duration (3:54.74) are also identical, so Tomahawk
can't distinguish them.

Should I start with a feature request for Tomahawk on Github?

>
> And obviously the Musicbrainz faction would also get it.
>
> What will be difficult for you is developing a large catalog of
> fingerprints. Have you checked in with the MB folks? They used to do
> fingerprinting. And it would be natural to use MB IDs for the
> underlying catalog.

Yeah, it will be trickier for songs that aren't available as a local
file. I haven't talked to MusicBrainz, but I do know they use AcoustIDs,
which run the Chromaprint algorithm under the hood.

https://wiki.musicbrainz.org/Fingerprinting

They link to several fingerprints for "King Kunta":

https://acoustid.org/track/c854f40f-aa37-42c0-9d95-13b5ff0c5328

>
> Happy to make introductions.

That would be great. Application support is important, and I would
really like to see XSPF become more widely adopted. Right now it seems
like most streaming services are allergic to interoperability.

>
> On Tue, Dec 22, 2015 at 4:00 PM, Nathaniel Beaver
> <[hidden email]> wrote:
>> I've extended XSPF to include Chromaprint fingerprints as described here:
>>
>> https://github.com/nbeaver/chromaprint-playlists
>>
>> Am I using the extension field correctly?
>>
>> Nathaniel Beaver
>> _______________________________________________
>> Playlist mailing list
>> [hidden email]
>> http://lists.xiph.org/mailman/listinfo/playlist
_______________________________________________
Playlist mailing list
[hidden email]
http://lists.xiph.org/mailman/listinfo/playlist
Loading...