Updated datasets or replication method?

replication
Tags: #<Tag:0x00007fe3d3ce29f8>

#1

Are there plans to update the publicly available AcousticBrainz data, or make replication available like it is for MusicBrainz?

I’m working on a project to listen to live audio and determine which of a couple candidate recordings it matches, and to align my projects internal clock to the beats_positions data from AcousticBrainz.

My early prototype is using a local MusicBrainz server and search server to get candidates from song metadata, but it’s having to pull from the AcousticBrainz www server to get low-level data. I’d rather not do this, for both performance and rate-limiting purposes (my normal use case will be to get 1 - 10 low-level datasets all at once, about once every 3-5 minutes).

For testing I can cache my test songs data offline, but going forward will there be a way to replicate the DB?


#2

Hi!
This is my fault, sorry. There has been some active work by some contributors to update the dump code in acousticbrainz, and I still haven’t merged it :frowning:


These are definitely on my list of things to do, but experience has taught me not to promise when it’ll be fixed by, because the chances that I’ll miss any deadline are very high!

Sorry that this is causing problems for you - hopefully I’ll be able to get it working for you as soon as I can.


#3

No worries, and thanks for the update. I can develop against the older data, just wanted to make sure it wasn’t going to be a dead-end long term.


#4

Would be helpful if this could be done because it is only possible to get details one recording at a time which is quite limiting when dealing with releases, especially since I need to do a low and high level call to get required data, see previous request AcousticBrainz datadumps

Also the low level call returns a lot of data but the only thing I actually need is BPM, being able to decide what you return may reduce bandwidth costs since I expect most users dont use the majority of data returned.