MusicBrainz is a MetaBrainz project that aims to create a collaborative music database that is similar to the freedb project. MusicBrainz was founded in response to the restrictions placed on the Compact Disc Database (CDDB), a database for software applications to look up audio CD information on the Internet. MusicBrainz has expanded its goals to reach beyond a CD metadata (this is information about the performers, artists, songwriters, etc.) storehouse to become a structured online database for music.[3][4]
MusicBrainz captures information about artists, their recorded works, and the relationships between them. Recorded works entries capture at a minimum the album title, track titles, and the length of each track. These entries are maintained by volunteer editors who follow community written style guidelines. Recorded works can also store information about the release date and country, the CD ID, cover art, acoustic fingerprint, free-form annotation text and other metadata. As of October 2023[update], MusicBrainz contains information on roughly 2.2 million artists, 3.9 million releases, and 30.4 million recordings.[5] End-users can use software that communicates with MusicBrainz to add metadata tags to their digital media files, such as ALAC, FLAC, MP3, Ogg VorbisorAAC.
MusicBrainz allows contributors to upload cover art images of releases to the database; these images are hosted by Cover Art Archive (CAA), a joint project between Internet Archive and MusicBrainz started in 2012. Internet Archive provides the bandwidth, storage and legal protection for hosting the images, while MusicBrainz stores metadata and provides public access through the Web and via an API for third parties to use. As with other contributions, the MusicBrainz community is in charge of maintaining and reviewing the data.[6] Until May 16, 2022,[7] cover art was also provided for items on sale at Amazon.com and some other online resources, but CAA is now preferred, because it gives the community more control and flexibility for managing the images. As of October 2023[update], over 4.6 million images exist in the archive.[8]
Besides collecting metadata about music, MusicBrainz also allows looking up recordings by their acoustic fingerprint. A separate application, such as MusicBrainz Picard, is used to do this.
In 2000, MusicBrainz started using Relatable's patented TRM (arecursive acronym for TRM Recognizes Music) for acoustic fingerprint matching. This feature attracted many users and allowed the database to grow quickly. However, by 2005 TRM was showing scalability issues as the number of tracks in the database had reached the millions. This issue was resolved in May 2006 when MusicBrainz partnered with MusicIP (now AmpliFIND), replacing TRM with MusicDNS.[9] TRMs were phased out and replaced by MusicDNS in November 2008.
In October 2009 MusicIP was acquired by AmpliFIND.[10] Sometime after the acquisition, the MusicDNS service began having intermittent problems.[citation needed]
Since the future of the free identification service was uncertain, a replacement for it was sought. The Chromaprint acoustic fingerprinting algorithm, the basis for AcoustID identification service, was started in February 2010 by a long-time MusicBrainz contributor Lukáš Lalinský.[11] While AcoustID and Chromaprint are not officially MusicBrainz projects, they are closely tied with each other and both are open source. Chromaprint works by analyzing the first two minutes of a track, detecting the strength in each of 12 pitch classes, storing these eight times per second. Additional post-processing is then applied to compress this fingerprint while retaining patterns.[12] The AcoustID search server then searches from the database of fingerprints by similarity and returns the AcoustID identifier along with MusicBrainz recording identifiers, if known.
In December 2004, the MusicBrainz project was turned over to the MetaBrainz Foundation, a non-profit group, by its creator Robert Kaye.[15] On 20 January 2006, the first commercial venture to use MusicBrainz data was the Barcelona, Spain-based Linkara in their『Linkara Música』service.[16]
On 28 June 2007, BBC announced that it had licensed MusicBrainz's live data feed to augment their music web pages. The BBC online music editors would also join the MusicBrainz community to contribute their knowledge to the database.[17]
On 28 July 2008, the beta of the new BBC Music site was launched, which publishes a page for each MusicBrainz artist.[18][19]
Picard identifies audio files and compact discs by comparing either their metadata or their acoustic fingerprints with records in the database.[20] Audio file metadata (or "tags") are a means for storing information about a recording in the file. When Picard identifies an audio file, it can add new information to it, such as the recording artist, the album title, the record label, and the date of release.[21]
ListenBrainz is a free and open source project that aims to crowdsource listening data from digital music and release it under an open license.[22] It is a MetaBrainz Foundation project tied to MusicBrainz. It aims to re-implement Last.fm features that were lost following that platform's acquisition by CBS.[23][24]
ListenBrainz takes submissions from media players and services such as Music Player Daemon, Spotify, and Rhythmbox in the form of listens. ListenBrainz can also import Last.fm and Libre.fm scrobbles in order to build listening history. As listens are released under an open license, ListenBrainz is useful for music research for industry and development purposes.[25][26][27][28][29]
^Vigliensoni, Gabriel; Fujinaga, Ichiro (23 October 2017). "The Music Listening Histories Dataset". Proceedings of the 18th International Society for Music Information Retrieval Conference. Suzhou, China: ISMIR: 96–102. doi:10.5281/zenodo.1417499. Retrieved 17 February 2024.
