MusicBrainz Picard


MusicBrainz Picard is a free MP3 file tagger that can identify track information by comparing an audio track’s “unique digital thumbprint” to the information in the MusicBrainz database. It is based on a tagging concept that is organized around album or release information for groups of tracks (and will identify individual tracks within that framework).

Imagine that you have a ripped album where the tracks are named “track1, track2, etc”, with no tags or metadata whatever. Imagine further that you do not even know the album or artist information, and do not recognize the songs. In this case it is not possible to use metadata management apps such as Mp3Tag or The Godfather or a similar program to identify the audio files; MusicBrainz Picard, however, is based on completely different technology and most likely will be able to get the information and apply a full profile of tags to your files. To test this I renamed some tracks so that the names contained no information and stripped all the tags, and, sure enough, Picard was able to identify and tag them. Here are some notes on this software:

  • The technology: MusicBrainz Picard creates a patented “unique digital thumbprint” for audio files that it processes and compares it to the MusicBrainz database in order to identify it. This database is completely user generated and maintained (you can get your own account and contribute information); as of May 2007 contained information about 306,369 artists, 483,324 albums, and 5.7 million tracks.
  • The interface: consists of 2 panels and is mostly simple and straightforward. Once you add a folder or tracks, Picard will display all tracks (recursively including those in subfolders) in one big list. The bottom panel will display the metadata and artwork for the selected track including the original metadata as well as the results of any online database scan for that track of.
  • How it works: Clicking ’cluster’ on the toolbar will instantly create subfolders that contain all the tracks that belong together in albums, such that the left panel is divided to tracks in clusters (i.e. albums) and “unmatched files”. The right panel will display information downloaded from the online database. Highlighting any entry in the left panel and selecting “scan” will result in a comparison with online data. Once you have had a chance to look over the new data you can right click “save” on the right hand panel to save info to disk. Discrepancies between existing and downloaded metadata are noted by an asterisk next to the downloaded album title as well as a small square green icon, making it easy to troubleshoot large library scans at a glance.
  • Album art: is the one area where MusicBrainz Picard falters, as album art does not seem to be provided (or perhaps will be supported in a future release?). Curiously, clicking on the blank album art thumbnail will produce an Amazon search for the album and display the album art in a browser window.
  • Supported formats: MP3, (ID3), Ogg Vorbis, FLAC, Speex (Vorbis comment),Musepack, WavPack (APEv2 tag), MP4, Windows Media Audio.

The verdict: this is simply a fantastic program that will help you find accurate information for all of your music, and will most likely (finally) fix those tracks in your library whose tag information you had all but given up on. The information in the MusicBrainz data base is not always 100% correct but comes close, and some of your tracks will simply not be identified (these you can actually add to the database yourself). Also keep in mind that this software is in Alpha as of this writing (although after continual use over 48 hours I have yet to encounter a bug).

I might also add that if you are processing large libraries of files where the metadata is available and correct (say for renaming purposes) then you are better off using Mp3Tag. For album art retrieval you would also use Mp3Tag or MediaMonkey, which I found to be the most effective for this task.

Version tested: 0.9.0 Alpha 14

Compatibility: WinAll, Linux (there are a number of programs that access the MusicBrainz database for Mac; see the MusicBrainz site for more info).

Go to the MusizBrainz Picard page to get the latest version (approx 6.51 megs). Also visit the Musicbrainz home page.