ListenBrainz tools for matching metadata to and from MusicBrainz.
Project description
ListenBrainz matching tools
In the ListenBrainz project we've created a number of tools and techniques that allowed us to create the MBID Mapping sub-project. This project takes an artist name and a recording (track) name and attempts to find the best matching recording in MusicBrainz.
This library will contain code to facilitate matching non-MusicBrainz metadata to MusicBrainz (tag a music collection) and from MusicBrainz to non-MusicBrainz metadata (content resolution).
Metadata cleaner
One component in this toolkit is the metadata cleaner, which removes (often) useless gunk from a metadata recording string:
"Tabula Rasa (feat. Lorraine Weiss)" -> "Tabula Rasa"
"Don't Give up - 2001 remaster" -> "Don't Give up"
This library also supports artist name cleaning:
"Amy Winhouse, Lorraine Weiss" -> "Amy Winehouse"
When a piece of metadata fails to be identified the metadata should be run through the metadata cleaner in an effort to remove this useless gunk that might be preventing a match. If the cleaner sucessfully removes gunk, the cleaned metadata should be looked up again, hoping that a match will be found this time.
To use this class, instantiate the MetadataCleaner class and call the clean_recording or clean_artist methods:
from lb_matching_tools.cleaner import MetadataCleaner
mc = MetadataCleaner()
mc.clean_recording("Tabula Rasa (feat. Lorraine Weiss)")
mc.clean_artist("Amy Winhouse, Lorraine Weiss")
Installation
To install this python module, simply do:
pip install .
Development
Create virtual environment
python -m venv .venv
Install dependencies
pip install .
Running tests
To run the internal unit tests, run:
pip install .
./test.py
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for lb-matching-tools-2024.1.30.1.tar.gz
Algorithm | Hash digest | |
---|---|---|
SHA256 | 6472a035edcca7da1cb88b463b1f8377943cdf39b5224ced680387223f873a8c |
|
MD5 | b60f88f4daec4acd3be30716e1a535d8 |
|
BLAKE2b-256 | a4545186e8be62ac2791b0ee6982ae05b8cf4404db8d57d59fa16ccd6c938b0b |
Hashes for lb_matching_tools-2024.1.30.1-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 50aa957ee8f64dbcac6b738e3707a26a05e14af25f18ea0dce6b51c388cca3f0 |
|
MD5 | 404803885d68c0bc63dc6f64692c437c |
|
BLAKE2b-256 | 17c4565d2124cf73e5683758818c3a7767634c8e8a5a8e67bbc7a0da38f39f0c |