A 24/7 music bot for Discord that pauses when nobody is listening.
To self-host this bot, you'll need the following:
- git
- node.js (Version 12 or higher is required.)
- ffmpeg
- A C/C compiler installed (Required to install
@discordjs/opus
.)
ffmpeg should be installed by default on Linux and MacOS, in case it isn't, install it with your package manager.
For Windows users:
For
ffmpeg
, head over to ffmpeg's official website to download the binary which will need to be added to your $PATH. If you don't know how to add folders to your $PATH, check out this guide.For
C/C compiler
, head over to Windows-Build-Tools on NPM and install the package globally.
In order to self-host this bot, first you'll need to clone this repository.
git clone https://github.com/moonstar-x/discord-music-24-7.git
Once cloned, proceed to install the dependencies:
npm ci --only=prod
Or, if you wish to install the devDependencies
:
npm install
After you have configured the bot, you can run it with:
npm start
Inside the config
folder, rename the file settings.json.example
to settings.json
and edit the file with you own Discord Token and other settings. If you don't have a Discord Token yet, you can see a guide on how to create it here.
Your file should look like this.
{
"token": "YOUR_DISCORD_TOKEN",
"prefix": "!",
"owner_id": "123123123",
"presence_type": "PLAYING",
"channel_id": "123213123",
"shuffle": true,
"pause_on_empty": true,
"owner_reporting": false,
"soundcloud_client_id": "",
"youtube_cookie": "",
"presence_status": "{status_icon} {song_name}"
}
You may also configure these options with environment variables. The settings set with the environment variables will take higher precedence than the ones in the config JSON file.
This table contains all the configuration settings you may specify with both environment variables and the JSON config file.
Environment Variable | JSON property | Required | Description |
---|---|---|---|
DISCORD_TOKEN | token |
Yes. | The bot's token. |
DISCORD_PREFIX | prefix |
No. (Defaults to: ! ) |
The bot's prefix used for the commands. |
DISCORD_OWNER_ID | owner_id |
No. (Defaults to: null ) |
The ID of the bot's owner. This is only necessary if owner_reporting is set to true . |
DISCORD_PRESENCE_TYPE | presence_type |
No. (Defaults to: PLAYING ) |
The presence type that will be used. Can only be one of: PLAYING , LISTENING , WATCHING , STREAMING , or COMPETING . |
DISCORD_CHANNEL_ID | channel_id |
Yes. | The ID of the channel that will be auto joined by the bot. |
DISCORD_SHUFFLE | shuffle |
No. (Defaults to true ) |
Whether the bot should shuffle the queue on start-up. |
DISCORD_PAUSE_ON_EMPTY | pause_on_empty |
No. (Defaults to true ) |
Whether the bot should pause the music when the channel is empty. |
DISCORD_OWNER_REPORTING | owner_reporting |
No. (Defaults to false ) |
Whether the bot should send DMs to the owner in case a command runs into an error. This is very unlikely with the amount of commands there is available. |
DISCORD_SOUNDCLOUD_CLIENT_ID | soundcloud_client_id |
Only required if queue contains a SoundCloud URL. | Your SoundCloud Client ID. To find your Client ID, check out this guide. |
DISCORD_YOUTUBE_COOKIE | youtube_cookie |
Preferable to use if your bot is hosted in a network with a static public IP. This will authenticate YouTube requests with your cookie. Tries to avoid the 429 Too Many Requests error. | Your YouTube cookie. To find your YouTube cookie, check out this guide. |
DISCORD_PLAYING_STATUS | playing_status |
No. (Defaults to {status_icon} {song_name} ) |
A customized playing status. Check customizing status for more information on how to use this. |
DISCORD_INTERMISSION_INTERVAL | intermission_interval |
No. (Defaults to null ) |
The number of songs that are played before an intermission audio is played. Check intermission audios for more information. |
- To see how to find the IDs for users or channels, you can check out this guide.
- If you don't have a Discord token yet, you can see a guide on how to get one here.
You can customize the bot's presence status message. This status is updated every time the bot plays or pauses music playback or when a new song is being played.
You can add template keys wrapped in braced {}
to aggregate dynamic information to your status message.
The following keys are supported:
{song_name}
: The current song being played.{song_source}
: The source of the song being played. This is added by the Provider being used and can have values such as:YT
,SC
,LOCAL
orGENERIC
.{status_icon}
: The icon of the status for the player. This corresponds to the stop, paused and playing icons.{song_index}
: The index of the current song in the queue.{queue_size}
: The size of the queue.
You can add how many of these you want into your status message. You can use as an example: {status_icon} [{song_index}/{queue_size}]
which will turn into something like ► [3/20]
.
Intermission audios try to emulate ads on a radio station. The intermission_interval
setting represents the number of songs that will be played before an intermission audio is played. This means that every X number of songs in the queue, an intermission audio will be played.
This setting needs to be a positive number greater or equal to 1.
If you set the intermission_interval
to null
in your config, no intermissions will be played. If you do add an intermission_interval
,
the bot will generate a folder named local-intermissions
and a file named intermissions.txt
inside the data
folder. Both of these act the same way as local-music
and queue.txt
, which means that intermission audios
can be added as local .mp3
, .opus
or .m4a
files inside the local-intermissions
folder and audio URLs can be added into the intermissions.txt
file.
If you set an intermission_interval
, you will need to add at least one file inside the local-intermissions
folder or add at least an audio URL inside intermissions.txt
, otherwise the bot will crash.
If you don't wish to use intermissions, simply leave intermission_interval
set to null
.
Keep in mind that since an intermission audio is not being pulled from the music queue, the {song_index}
status key will not be incremented.
Currently, the presence status message update does not differentiate between a song being played from the queue and an intermission audio being played.
You can start a container with the bot's image by running:
docker run -it -e DISCORD_TOKEN="YOUR DISCORD TOKEN" -e CHANNEL_ID="YOUR CHANNEL ID" -v "/local/folder/for/data":"/opt/app/data" moonstarx/discord-music-24-7:latest
Check configuration to see which environment variables you can use.
The following volumes can be used:
/opt/app/config
: The config folder for the bot, here you can use thesettings.json
file to configure the bot if you don't want to use environment variables./opt/app/data
: The data folder for the bot. Here you can find thequeue.txt
file containing all the song URLs for the bot to play. There will also be alocal-music
folder to insert your MP3, OPUS or M4A files to play.
Here's a list of all the commands for the bot:
Command | Alias | Description |
---|---|---|
!skip | !s | Skip the current song being played. |
!help | !h | Display a help message with all the available commands. |
The bot will automatically start playing music in the configured channel, it will also automatically pause when nobody is listening to the music to save bandwidth (if pause_on_empty
or PAUSE_ON_EMPTY
are enabled).
This bot is designed to work only on one server at a time. Inviting your bot to multiple servers will yield unexpected behavior.
The music that is played is stored in the queue.txt
inside the data
folder. You can add the music you want the bot to play in this file. Any links must begin with https://
or http://
.
Additionally, you can play local music by inserting .mp3
, .opus
or or .m4a
files inside the data/local-music
folder. Make sure the files are properly ID3 tagged so the bot can get the artist and song name from the metadata.
The bot can play music from the following URLs:
youtube.com
youtu.be
soundcloud.com
Playlists are not officially supported yet. You may find that they may work but generally only the first song will be pulled from them.
You can also play radio streams, as long as the stream is of the following types:
audio/aac
audio/mpeg
audio/ogg
audio/opus
audio/wav
Pausing/Resuming may take a bit longer with these streams.
If you're reaching 429: Too Many Requests errors in the console, then most likely you're being rate limited by YouTube because your connection is not authenticated.
Inside Configuration you'll find a parameter called youtube_cookie
, this is where you'll paste the cookie extracted from your YouTube login.
- Head over to YouTube and login if you haven't already.
- Then, open the developer inspector (Ctrl Shift I or Cmd Option I) and open the Network tab. Refresh the page if you're not getting a list of requests. (Make sure to select All* in the filter tab).
- Look for a request named
www.youtube.com
and check it's headers. UnderResponse Headers
you will find the headerset-cookie
, this will contain a directive calledexpires
, keep this date in mind because your cookie will be invalidated after this date and you may need to update the settings file manually prior to this date. (Currently looking for a way to avoid this.) - Inside the same request, look for the
Request Headers
and copy the content of thecookie
header. This is the cookie that you'll need to paste inside thesettings.json
file or set as your YOUTUBE_COOKIE environment variable.
This solution has varied results, it works for some people, it doesn't for others.
Unfortunately, Repl.it does not support ffmpeg anymore which means that this bot will not work on this.
This bot was made by moonstar-x with contributions from zackradisic (Soundcloud support).