beaTunes News

Friday, June 30, 2017

AcousticBrainz Plugin Available Now

If you haven't heard of AcousticBrainz yet, it's a fantastic project that "aims to crowd source acoustic information for all music in the world and to make it available to the public". Crowd source means that the project needs people like you and me to run a so-called feature extractor through their music libraries and submit the results to the AcousticBrainz database. There are multiple ways to do this—e.g. you can use the tools available on the AcousticBrainz website or the MusicBrainz Picard Tagger. And as of today, there is another way.

beaTunes 5 offers a new plugin that runs the feature extractor and submits the results for you. You can install it via the plugin preferences (see screenshot above, select plugin, click on the + button). Once installed, it's packaged as an analysis task, which means, that it shows up at the bottom of the available tasks whenever you run an analysis. Just check the box and beaTunes will run it.

Please note that running the extractor is not a quick thing. It does a lot of things very thoroughly, so analysis will take a long time. But you'll be doing a good thing! :-)

Thank you for your contribution. Among others, beaTunes uses AcousticBrainz data internally and indirectly benefits from your submissions.

The source code for the plugin is available on GitHub.

Labels: ,

7 Comments:

Blogger Unknown said...

Is there a way to flag in Beatunes which files already have an Acoustic Brainz submittal or ID? I don't want to waste time re-analyzing files that already are in the database.

June 30, 2017 at 8:58:00 PM EST  
Blogger beaTunes said...

No, sorry. That's not possible right now.

July 1, 2017 at 2:03:00 AM EST  
Blogger Paul Taylor said...

If the songs are analysed but do not have a Musicbrainz recording id then the results will be ignored by AcousticBrainz so although it can help with local results it will not contribute to the AcousticBrainz project, so I assume the plugin should only be run after Beatunes has done song identification.

July 20, 2017 at 2:54:00 AM EST  
Blogger beaTunes said...

Not entirely true. The plugin code attempts to lookup the MBID if it cannot find an embedded one. See https://github.com/beatunes/plugin-samples/blob/master/abzsubmit/src/main/java/com/beatunes/acousticbrainz/AcousticBrainzSubmit.java#L320
If no MBID can be found, the plugin does not run the analysis for the given song (https://github.com/beatunes/plugin-samples/blob/master/abzsubmit/src/main/java/com/beatunes/acousticbrainz/AcousticBrainzSubmit.java#L124).

July 20, 2017 at 2:42:00 PM EST  
Blogger beaTunes said...

Sorry, I should have made the links clickable:

MBID lookup

Skip logic

July 20, 2017 at 2:43:00 PM EST  
Blogger Unknown said...

Maybe I don't understand the idea of the project but none of my songs have MusicBrainz IDs... so none of them submit information.

Analysis: Failed to submit 'Whatever Track' to AcousticBrainz. Unable to find MusicBrainz ID. (id=-7570423991724375826)

How do new tracks get added to MusicBrainz ID if unrecognized tracks are all rejected?

September 6, 2017 at 1:36:00 PM EST  
Blogger beaTunes said...

beaTunes will attempt to lookup a MusicBrainz ID for your songs online (in its own database). If it fails, you see that message. Note that not all songs have MusicBrainz IDs. You can of course check manually, whether they are on http://musicbrainz.org

September 6, 2017 at 1:48:00 PM EST  

Post a Comment

<< Home