• Server address: The domain name for the MusicBrainz database server used by Picard to get details of your music. Default value: musicbrainz.org (for the main MusicBrainz server).
  • Port: The port number for the server. Default value: 80 (for the main MusicBrainz server).
  • Username: Your MusicBrainz website username, used to submit acoustic fingerprints, retrieve and save items to your collections and retrieve personal folksonomy tags.
  • Password: Your MusicBrainz website password.
  • Automatically scan all new files: Check this box if you want Picard to scan each music file you add and look for an AcoustID fingerprint. This takes time, but may be helpful for you and MusicBrainz. Leave it unchecked if you don't want Picard to do this scan automatically. In any case, you can direct Picard to scan a particular music file at any time using the Scan button in the toolbar.
  • Ignore MBIDs when loading new files: If you disable this option Picard will not use MusicBrainz identifiers (MBIDs) stored in the files to automatically load the corresponding MusicBrainz release and match the loaded file to the correct track.
  • Check for updates during start-up: Check this box if you want Picard to check for program updates automatically. Leave it unchecked if you don't want Picard to do this check automatically. In any case, you can have Picard check for program updates at any time using the Help --> Check for update command in the drop-down menu.
  • Days between checks: This option allows you to limit the automatic update checking by selecting the interval, in days, between checks. Set this to 1 if you want to check daily, 7 for weekly checks, and so on. Note that this only applies if the Check for updates during start-up option is selected.
  • Updates to check: This option allows you to select which levels of update to check. Your options are:
    • Stable releases only
    • Stable and Beta releases
    • Stable, Beta and Dev releases
    For example, if you subscribe to Stable releases only you will not be notified if a new Beta or Dev release is issued.

Note: The update settings and "Check for update" function may not be available when Picard is distributed as a package. In that case, the user should check with the maintainer of the package to determine when an update is available.


  • Translate artist names to this locale where possible: When checked, Picard will see whether an artist has an alias for the selected locale. If it does, Picard will use that alias instead of the artist name when tagging. When "English" is the selected locale, the artist sort name (which is, by Style Guideline, stored in Latin script) is used as a fallback if there is no English alias.
  • Use standardized artist names: Check to only use standard Artist names, rather than Artist Credits which may differ slightly across tracks and releases.
    Note: If the "translate artist names" option above is also checked, it will override this option if a suitable alias is found.
  • Use standardized instrument and vocal credits: Check to only use standard names for instruments and vocals in performer relationships. Uncheck to use the instruments / vocals as credited in the relationship.
  • Convert Unicode punctuation characters to ASCII: Converts Unicode punctuation characters in MusicBrainz data to ASCII for consistent use of punctuation in tags. For example, right single quotation marks (’) are converted to ASCII apostrophes ('), and horizontal ellipses (…) are converted to three full stops (.).
  • Use release relationships: Check to retrieve and write release-level relationships to your files, e.g. URLs, composer, lyricist, performer, conductor, DJ mixer, etc. (You must have this enabled to use Picard to retrieve cover art)
  • Use track relationships: Check to write track-level relationships to your files, e.g. composer, lyricist, performer, remixer, etc.
  • Various artists: Choose how you want the 'Various Artists' artist spelled.
  • Non-album tracks: Choose how you want 'non-album tracks' (Recordings that do not belong to any Release) to be grouped.

Preferred Releases

  • Preferred release types: Adjust the sliders for various release types to tweak how likely Picard is to match a file or cluster to releases of various types. You can use this to decrease the likelihood of Picard matching a file or album to a Compilation or Live version, for example.
  • Preferred release countries: Add one or more countries into the list to make Picard prefer matching clusters/files to releases from the chosen countries. This list is also used to prioritise files in the "Other Releases" context menu.
  • Preferred release formats: Add one or more formats into the list to make Picard prefer matching clusters/files to releases of the specified format. This list is also used to prioritise files in the "Other Releases" context menu.


  • Use genres from MusicBrainz: Use genres provided by MusicBrainz and save them to the genre tag.
  • Fall back on album's artists genres if no genres are found for the release or release group: If there is no genre set for the release or release group on MusicBrainz use the genre of the album artist instead.
  • Only use my genres: Check to only write genres you personally have submitted to MusicBrainz. You'll need to set your username and password to use this feature.
  • Use folksonomy tags as genres: Check to use all folksonomy tags to set the genre. Otherwise only the tags considered by MusicBrainz to be proper genres will be used.
  • Minimal genre usage: Choose how popular the genre must be before it is written by Picard. Default: 90%. Lowering the value here will lead to more, but possibly less relevant, genres in your files.
  • Maximum number of genres: Choose how many genres to use. Default: 5. If you only want a single genre, set this to 1.
  • Join multiple genres with: Select which character should be used to separate multiple genres.
  • Genres or folksonomy tags to include or exclude: One expression per line, case-insensitive. You can use the "Playground" text field to enter some genres and test the rules you have setup. Genres that will be excluded will be marked red, included genres will be marked green.


    Lines not starting with - or + are ignored.


    Strict filtering:

    Exclude exactly word by prefixing it with -:


    Include exactly word, even if another rule would exclude it, by prefixing it with +:


    Wildcard filtering:

    Exclude all genres ending with word:


    Include all genres starting with word:


    Exclude all genres starting with w and ending with rd:


    Regular expressions filtering (Python re syntax):

    Exclude genres starting with w followed by any character, then r followed by at least one d:



  • Enable track ratings: Check to write track ratings to your files.
  • Submit ratings to MusicBrainz: Check to submit ratings to MusicBrainz. The tracks will be rated with your account.


  • Write tags to files: Uncheck to disable Picard from writing data. Picard may still move/rename your files according to your settings.
  • Preserve timestamps of tagged files: If checked, does not update the Last Modified date and time of your music files when it writes new tags to them.

Before Tagging

  • Clear existing tags: Checking this will remove all existing metadata and leave your files with only MusicBrainz metadata. Information you may have added through another media player such as genre, commentsor ratingswill be removed.
  • Remove ID3 tags from FLAC files: Check to remove ID3 tags from FLAC files – Vorbis Comments are recommended for FLAC files. Picard will write Vorbis Comments to FLACs regardless of this setting.
  • Remove APEv2 tags from MP3 files: Check to remove APEv2 tags from MP3 files – ID3 is recommended for MP3s. Picard will write ID3 tags to MP3s regardless of this setting.
  • Preserve these tags from being cleared or overwritten with MusicBrainz data: This is an advanced option: If you have tags which you need to preserve, enter their names here to stop Picard from overwriting them.

Tag Compatibility

  • ID3v2 version: Although id3v2.4 is the latest version, its support in music players is still lacking. Whilst software such as foobar2000 and MediaMonkeyhave no problem using version 2.4 tags, you will not be able to read the tags in Windows Explorer or Windows Media Player (in any Windows or WMP version, including those in Windows 8.1). Apple iTunes is also still based in id3v23, and support for id3v24 in other media players (such as smartphones) is variable. Other than native support for multi-valued tags in v2.4, the Picard Tag Mapping will show you what you lose when choosing v2.3 instead of v2.4.
  • ID3v2 text encoding: The default for version 2.4 is UTF-8, the default for version 2.3 is UTF-16. Use ISO-8859-1 only if you face compatibility issues with your player.
  • Join id3v23 tags with: As mentioned above, id3v23 does not support multi-value tags, and so Picard flattens these to strings before saving them to id3v23 tags. This setting defines the string used to separate the values when flattened. Use '; ' for the greatest compatibility (rather than '/' since tags more often contain a / than a ;) and for the best visual compatibility in Picard between id3v23 and other tagging formats.
  • Save iTunes compatible grouping and work: Save the tags grouping and work so that they are compatible with current iTunes versions. Without this option grouping will be displayed in iTunes as "work name" and work will not be available. See the Picard Tag Mapping for details.
    Note: For other players supporting grouping and work you might need to disable this option. MusicBee is one example for this.
  • Also include ID3v1 tags in the files: Not recommended at all. ID3v1.1 tags are obsolete and may not work with non-latin scripts.
  • AAC / AC3 files: Picard can save APEv2 tags to pure AAC or AC3 files, which by default do not support tagging. APEv2 tags in AAC or AC3 are supported by some players (e.g. foobar2000 or MusicBee), but players not supporting AAC or AC3 files with APEv2 tags can have issues loading and playing those files. Most often they display a wrong duration, causing issues on track change. To deal with this you can choose whether to save tags to those files:

    • Save APEv2 tags: Picard will save APEv2 tags to the files.
    • Do not save tags: Picard will not save any tags to the files, but you can still use Picard to rename them. By default existing APEv2 tags will be kept in the file.
    • Remove APEv2 tags: If you have "Do not save tags" enabled checking this option will cause Picard to remove existing APEv2 tags from the file on saving.

    Regardless of how you have configured saving tags Picard will always read existing APEv2 tags in AAC or AC3 files.

Cover art

You must enable "Option / Metadata / Use release relationships" for Picard to be able to download cover art from MusicBrainz cover art relationships.


  • Embed cover images into tags: Enables images to be embedded directly into your music files. Whilst this will use more storage space than storing it as a separate image file in the same folder, some music players will only display embedded images and don't find the separate files.
  • Only embed a front image: Embeds only a front image into your music files. Many music players will only display a single embedded image, so embedding additional images may not add any functionality.
  • Save cover images as separate files: In the file name mask you can use any variable or function from Picard Tags and Picard Scripting. The mask should not contain a file extension; this is added automatically based on the actual image type. The default value is cover. If you change this to folder, Windows will use it to preview the containing folder.
    In addition to scripting variables already available for a track you can use the following cover art specific variables:
    • coverart_maintype: The primary type (e.g. front, medium, booklet etc.). For front images this will always be "front".
    • coverart_types: Full list of all types assigned to this image.
    • coverart_comment: The cover art comment.
  • Overwrite the file if it already exists: Check this to replace existing files. This is especially recommended if trying to write "folder" previews for Windows.

Cover Art Providers

Picard can download Cover Art from a number of sources, and you can choose which sources you want Picard to download cover art from. You can activate more than one provider and choose the order in which the providers are used. Picard will try the providers from top to bottom and the first provider which provides images will be used.

  • Cover Art Archive: The Cover Art Archive (CAA) is MusicBrainz own archive of cover art in cooperation with the Internet Archive (archive.org). If art is available there, the Cover Art Archive is the most comprehensive database of cover art (front covers, back covers, booklets, CDs etc.).
  • CAA Release Group: This provider uses the Cover Art Archive cover image assigned to the release group. This is usually the image that best describes the release group as a whole or the image with the best visual quality, but is not necessarily the exact cover of the release your are tagging. This provider is a good choice if you care more about visual quality then having an exact representation of your release. It is also a good fallback for the Cover Art Archive provider.
  • Sites on the whitelist: See Style/Relationships/URLs/Cover_art_whitelist
    Note: CD Baby and other whitelist sites are no longer being used by MusicBrainz for new Cover Art.
  • Local Files: Load cover art from local files. The file names to load can be configured in the Local Files provider options.

In addition to the built-in cover art providers described above, additional cover art providers can be installed as plugins.

  • Amazon: Amazon often has cover art when other sites don't, however whilst this art is almost always for the correct Artist/Album, it may not be the absolute correct cover art for the specific Release that you have tagged your music with.
    Note: The Amazon cover art provider was built-in in Picard 2.1.3 and earlier versions. For later versions it needs to be installed as a separate plugin.
  • fanart.tv: Uses cover art from fanart.tv, which focuses on cover art with high visual quality. This provider provides cover art representative for the release group and not the individual release.
  • TheAudioDB: Uses cover art from TheAudioDB, which focuses on cover art with high visual quality. This provider provides cover art representative for the release group and not the individual release.

Cover Art Archive

In this section you can decide which types of cover art you would like to download from the Cover Art Archive, and what quality (size) you want to download. Obviously, the better the quality, the larger the size of the files.

Most music players will display only one piece of cover art for the album, and most people select Front (cover) for that.

When selecting the cover art image types, you can select both the types to include and exclude from the download list. CAA images with an image type found in the 'Include' list will be downloaded and used unless they also have an image type found in the 'Exclude' list. Images with types found in the 'Exclude' list will never be used. Image types not appearing in the 'Include' or 'Exclude' lists will not be considered when determining whether or not to download and use a CAA image.

Since Picard 1.3, you can also decide to use the image from the release group (if any) if no front image is found for the release. In this case, the cover may not match the exact release you are tagging (eg. a 1979 vinyl front cover may be used in place of the Deluxe 2010 CD reissue).

Local Files

In this section you can configure the file names to be used by the Local Files cover art provider. The file names are defined using a regular expression. The default is to load files with the name cover, folder or albumart and the file extension jpg, png, gif or tiff (e.g. folder.jpg or cover.png).

File Naming

This page tells Picard whether it should move your audio files to a new directory when it saves metadata in them. One use for this is to keep your work organised: all untagged files are under directory A, and when Picard tags them it moves them to directory B. When directory A is empty, your tagging work is done. Check this box, and select a destination directory, if you want Picard to move files this way. Uncheck the box if you want Picard to leave the files under the same directory.

The Rename Files and Move Files options are independent. Rename Files refers to Picard changing file names typically based on artist and track names. Move Files refers to Picard moving files to new directories, based on a stated parent directory and sub-directories typically based on album artist name and release title. However, they both use the same "file naming string". Move files uses the portion up until the last '/'; rename files the part after that.

  • Rename files when saving: Check to let Picard change file and directory names of your files when it saves metadata in them, in order to make the file and directory names consistent with the new metadata.
  • Replace non-ASCII characters: Check to replace non-ASCII characters with their ASCII equivalent, e.g. á,ä,ǎ, with a; é,ě,ë, with e; æ with ae, etc. For more information on ASCII characters read the Wikipedia page on ASCII.
  • Replace Windows-incompatible characters: Check to replace Windows-incompatible characters with an underscore. Enabled by default on Windows with no option to disable.
  • Move files when saving: Choose a destination parent directory to move saved files to.
    • If you use the directory "." they will be moved relative to their current location. If they are already in some sort of folder structure, this will probably not do what you want!
  • Delete empty directories: Check to have Picard remove directories that have become empty once a move is completed. Leave unchecked if you want Picard to leave the source directory structure unchanged. Checking this box may be convenient if you are using the move files option to organise your work. An empty directory has no more work for you to do, and deleting the directory makes that clear.
  • Move additional files: Enter wildcard patterns that match any other files you want Picard to move when saving files, e.g. Folder.jpg, *.png, *.cue, *.log. Using default settings, when these additional files are moved they will end up in the release directory with your files. In a wildcard, *matches zero or more characters. Other text, like .jpg, matches those exact characters. Thus *.jpgmatches "cover.jpg", "liner.jpg", "a.jpg", and ".jpg", but not "nomatch.jpg2". Put spaces between wildcard patterns.
  • Name files like this: An edit box that contains a formatting string that tells Picard what the new name of the file and its containing directories should be, in terms of various metadata values. The formatting string is in Picard's scripting language where dark blue text starting with a "$" is a function name and names in light blue within "%" signs are Picard's tag names. Note that the use of a "/" in the formatting string means that everything before the string is a directory name, and everything after the last "/" becomes the file's name. The formatting string is allowed to have zero, one, or multiple, "/".


If you select a file or cluster in the Left side of the Picard screen and click Scan, Picard will invoke a program to scan the file and produce a fingerprint that can then be used to look up the file on MusicBrainz.

MusicBrainz currently supports only AcoustID (an Open Source acoustic fingerprinting system created by Lukáš Lalinský) but has previously supported TRM and MusicID PUID.

CD lookup

This is where you tell Picard which CD drive it should use for looking up CDs.


On Windows, Picard has a pulldown menu listing the various CD drives it has found. Pull down the menu and select the drive you want.


In macOS, this option is currently a text field. The device is usually /dev/rdisk1.

If that doesn't work, one way is to simply keep increasing the number (e.g. /dev/rdisk2) until it does work. A less trial and error method is to open Terminal and type mount. The output should include a line such as /dev/disk2 on /Volumes/Audio CD (local, nodev, nosuid, read-only). You need to replace /dev/disk with /dev/rdisk, so if, for example, it says /dev/disk2, you should enter /dev/rdisk2 in Picard's preferences.


In Linux, Picard has a pulldown menu like in Windows. If you're using an older version with a text field, you should enter the device name (typically /dev/cdrom) here.

Other platforms

On other platforms, the CD Lookup option is a text field and you should enter the path to the CD drive here.


Here you may enable/disable any of the plugins you have installed in Picard. Note that some plugins have their own option page which will appear under here.

For a list of plugins see Picard Plugins.

User interface

  • Show text labels under icon: Uncheck to make the toolbar a little smaller.
  • Allow selection of multiple directories: Check to bypass the native directory selector and use QT's file dialog since the native directory selector usually doesn't allow you to select more than one directory. This applies for the 'Add folder' dialog, the file browser always allows multiple directory selection.
  • Use advanced query syntax: Check to enable advanced query syntax parsing on your searches. This only applies for the search box at the top right of Picard, not the lookup buttons.
  • Show a quit confirmation dialog for unsaved changes: Check to show a dialog when you try to quit Picard with unsaved files loaded. This may help prevent accidentally losing tag changes you've made, but not yet saved.
  • Begin browsing in the following directory: By default, Picard remembers the last directory you loaded files from. If you check this box and provide a directory, Picard will start in the directory provided instead.
  • User interface language: By default, Picard will display in the language displayed by your operating system, however you can override this here if needed.


You can customize various colors used in the Picard user interface here.

Color name Description
Errored entity Color for files and other elements with errors on loading or saving
Pending entity Color for files and other elements queued up for processing
Saved entity Color for successfully saved files
Log view text (debug) Color for debug messages in the Error/Debug Log
Log view text (error) Color for error messages in the Error/Debug Log
Log view text (info) Color for informational messages in the Error/Debug Log
Log view text (warning) Color for warning messages in the Error/Debug Log
Tag added Color for newly added tags in the metadata view at the bottom
Tag changed Color for changed tags in the metadata view at the bottom
Tag removed Color for removed tags in the metadata view at the bottom

Top Tags

The tags specified here will always be shown in this order at the top of the metadata view (which shows the metadata of selected files or tracks in the lower section of the main window). This allows you to have the most important tags always on top of the list. Tags not listed here will be shown in alphabetical order below the top tags.


For scripting help see Picard Scripting and Picard Tags for variables available to script with.


  • Ignore file paths matching the following regular expression: You can specify patterns for files and folders to never load into Picard. E.g. if you set this to the regular expression \.bak$ any file ending in ".bak" will be ignored when loading files.
  • Ignore hidden files: If enabled hidden files and folders will not be loaded. This includes also any file or sub-folder inside a hidden folder.
  • Include sub-folders when adding files from folders: If enabled Picard will load all audio files in the selected folder and all its sub-folders. If disabled only audio files directly in the selected folder will get loaded.
  • Ignore track duration difference under this number of seconds: If a file differs from the length in the MusicBrainz database by less then the number of seconds configured here (default: 2 seconds) it will still be considered the same.
  • Ignore the following tracks when determining whether a release is complete: Missing tracks of the selected type (video, pregap, data or silence) will be ignored when checking a release for completeness. E.g. if "video" is selected here a release with a bonus video will be marked as complete if it has all the audio tracks matched with a file even if the video file is missing.
  • Tags to ignore for comparison: Tags in this list will not be considered when comparing the existing file metadata to the data loaded from MusicBrainz. A file where the only tag difference is with one of the tags in this ignore list will still be considered unmodified.


Web Proxy

If you need a proxy to make an outside connection you may specify one here.

Browser Integration

The browser integration allows you to load releases and recordings into Picard directly from the MusicBrainz website. Once you have opened musicbrainz.org in your browser from Picard the website will show the green tagger button Tagger next to releases and recordings, clicking on this button will load the corresponding release or recording into Picard.

  • Default listening port: The default port Picard will listen on for the browser integration. If the port is not available Picard will try to increase the port number by one until it finds a free port.
  • Listen only on localhost: By default Picard will limit access to the browser integration port to your local machine. Deactivating this option will expose the port on your network, allowing you to request Picard to load a specific release or recording via the network. This is used for example by the Picard Barcode Scanner Android app.

    Warning: Only enable this option when you actually need it and only on networks you trust. Exposing application ports via the network can open potential security holes on your system.


It is recommended for most users to not change these settings. However for advanced users, it allows you to tune the way Picard matches your files and clusters to to MusicBrainz releases and tracks.

  • Minimal similarity for file lookups: The higher then %, the more similar an individual file's metadata must be to MusicBrainz's metadata for it to be moved/matched to a release on the right-hand side.
  • Minimal similarity for cluster lookups: The higher then %, the more similar a cluster of files from the left-hand pane must be to a MusicBrainz release for the entire cluster to be moved/matched to a release on the right-hand side.
  • Minimal similarity for matching files to tracks: The higher then %, the more similar an individual file's metadata must be to MusicBrainz's metadata for it to be moved/matched to a release on the right-hand side.

If you have absolutely no metadata in your current files, and you are using Scan to match tracks, you may find you need to lower Minimal similarity for matching files to tracks in order to get Picard to match the files within a release. Otherwise you may find that Picard matches the track to a release but then is not sure which track is correct; and leaves it in an "unmatched files" group within that release.

As a general rule, lowering the percentages may increase the chance of finding a match at the risk of false positives and incorrect matches.