Skip to content
Go to file

Latest commit


Git stats


Failed to load latest commit information.
Latest commit message
Commit time


Let the crowd make its own mix without hassling you.

Think about a web-based, crowd-powered jukebox that uses your audio player of choice in background.


  • Local, searchable music database created from your audio files
  • Support 8 audio formats (see below for the list)
  • Support 11 audio players (see below for the list)
  • Responsive (can be used on mobile devices)
  • (Optional) Display the currently playing song
  • Two queuing mode (vote or immediate)
  • Time-based song submission restrictions
  • Internationalized & localized in 2 languages:
    • English (en)
    • French (fr)


  • Should work on any Python 3.x version. Feel free to test with another Python version and give me feedback
  • A uWSGI-capable web server (optional, but recommended)
  • One of the supported audio players (see the Supported audio players section below)


  1. Clone this repo somewhere
  2. pip install -r requirements.txt
  3. pybabel compile -d translations
  4. IMPORTANT: Other dependencies are needed regarding the audio player you'll use. Please refer to the table in the Supported audio players section below and install them accordingly using pip install <package> before continuing
  5. export (Windows users: set
  6. flask create_database (WARNING: don't re-run this command unless you want to start from scratch, it will wipe out all the data)
  7. flask index (this will index your songs. Don't forget to set the MUSIC_DIRS configuration parameter before, read below. Run flask index --help for the full list of arguments)


Copy the file to and fill in the configuration parameters.

Available configuration parameters are:

  • SECRET_KEY Set this to a complex random value

More informations about Flask config values can be found here.

  • TITLE If set to a string, will be used to replace the default app title (which is "CrowdMixer")
  • FORCE_LANGUAGE Force the lang to be one of the supported ones (defaults to None: auto-detection from the Accept-Language HTTP header). See in the features section above for a list of available lang keys
  • DEFAULT_LANGUAGE Default language if it cannot be determined automatically. Not taken into account if FORCE_LANGUAGE is defined. See in the features section above for a list of available lang keys
  • CACHE_THRESHOLD The maximum number of items the cache will store before it starts deleting some (see here for more configuration parameters related to Flask-Cache)
  • MUSIC_DIRS A list of absolute paths to directories containing songs (read below for the list of supported formats)
  • NOW_PLAYING_CACHE_TIME Number of seconds the "Now playing" information will be stored in the cache
  • MODE Submit mode that should be used. Can be either Immediate (song is queued immediately) or Vote (song is queued when a votes threshold is reached)
  • VOTES_THRESHOLD If MODE is equal to Vote: number of votes required to actually queue a song in the playlist
  • BLOCK_TIME Define the number of seconds a song that have just been queued is unavailable for submitting
  • REQUEST_LIMIT Define the minimum number of seconds users have to wait between each submit
  • SHOW_CURRENT_PLAYING Enable or disable the display of the currently playing song (support may vary following the audio player used, more information in the Supported audio players section below)
  • SONGS_PER_PAGE How many songs to display per page
  • PLAYER_TO_USE The audio player to use. Can be one of the ones in the table below, in the Supported audio players section
  • PLAYERS Self-explanatory audio players-specific configuration values. Change them if your audio player of choice (PLAYER_TO_USE) is requiring it (see the table below, in the Supported audio players section)

I'll let you search yourself about how to configure a web server along uWSGI.

Some audio players needs to be themselves configured, you'll find the configuration instructions below according to your audio player of choice.


You'll have to enable the remote controlling feature of Clementine before to use CrowdMixer. To do so:

  1. In the menu bar of Clementine, click Tools > Settings...
  2. In the Remote control tab, check the Use remote control checkbox
  3. At this point, you can either use the default parameters, or customize them. If so, don't forget to change the configuration values of Clementine accordingly in your
  4. Click on OK

Music Player Daemon

MPD, by design and by default, is only controllable via a TCP connection on *:6600. You can change this behavior in the /etc/mpd.conf configuration file (depending of your operating system). If so, don't forget to change the configuration values of MPD accordingly in your

For more information, please read the MPD man page: man mpd.conf or read the example configuration file.


You'll have to enable the web interface feature of VLC before to use CrowdMixer. To do so:

  1. In the menu bar of VLC, click Tools > Settings...
  2. On the bottom left of the Settings window, click on All in the Settings to display box
  3. In the left list box that has been shown, click on Interface > Main interface
  4. Check the Web checkbox
  5. In the left list box, click on Interface > Main interface > Lua
  6. In the Lua via HTTP > Password textbox, enter a password that will be required to remote control VLC. If you do not enter a password, remote control will not be possible for security reasons
  7. At this point, you can either save the settings, or customize them. For more information, please read the VLC documentation

Don't forget to change the configuration values of VLC according to your VLC settings in your


  • Standalone

Run the internal web server, which will be accessible at http://localhost:8080:


Edit this file and change the interface/port as needed.

  • uWSGI

The uWSGI file you'll have to set in your uWSGI configuration is The callable is app.

  • Others

You'll probably have to hack with this application to make it work with one of the solutions described here. Send me a pull request if you make it work.

How it works

This project is built on Flask (Python) for the backend which is using an SQLite database to persist data. The flask index command is used to index the songs with the help of the tinytag package. Those songs can then be browsed and submitted for playing using the web interface provided by Flask.

For more information about indexing, see the index() function in the file.

For more information about methods used to retrieve the currently playing song and to queue songs, see the file.

Supported audio file formats

CrowdMixer maintain its own music library database. The following audio file formats are supported by the CrowdMixer music indexer:

  • .mp3, .m4a
  • .ogg, .oga, .opus
  • .flac
  • .wma
  • .wav

Make sure your audio player of choice also support these, otherwise you'll get errors while queuing songs.

Supported audio players

CrowdMixer requires to be ran on the same computer that is running your audio player of choice.

Name "Now playing" supported? Configuration value Additional PyPI dependencies Needs additional configuration in
AIMP Aimp pyaimp
Audacious Audacious psutil
Clementine Clementine protobuf
foobar2000 Foobar2000 psutil
MediaMonkey MediaMonkey psutil
MusicBee MusicBee psutil
Music Player Daemon Mpd python-musicpd
Rhythmbox Rhythmbox
VLC Vlc requests
Winamp Winamp psutil
XMMS2 Xmms2 xmmsclient

The following audio players cannot be supported by CrowdMixer for technical reasons:

  • Windows Media Player
  • iTunes

End words

If you have questions or problems, you can submit an issue.

You can also submit pull requests. It's open-source man!

You can’t perform that action at this time.