Discord Soundmoji

What does this bot do?

It adds sounds to your emojis (including the custom ones) and words.

Really? How does it work?

The users have to connect to a voice channel. If they type a keyword, it'll be caught by the bot and a specified sound will be played. The keywords are specified with regular expressions, so you can have a couple of words for one sound. The possibilities are endless. For example: /lol|haha|heh/i will be triggered if an user writes lol, haha or heh (and it's case insensitive, so LOL will also work). You can also have a couple of sounds specified. For example lol can play back either laugh1.mp3 or laugh2.mp3, chosen randomly.

(Current) limitations

• The bot won't play a couple sounds at once. It can stop playing a sound that's currently being played and play another sound.
• The bot plays only one sound from a message. Messages like first_keyword another_keyword won't queue two sounds.
• Sounds aren't being queued.
• The bot can be connected to many servers, but only one voice channel, which means every server will have to set up their own bot.

Adding your own sounds

Directory structure

You have to create a file named snd.yml. In this file you should define the directories, which will contain the sound files. Each of these directories has to contain a snd_pack.yml file with keywords and the appropriate sound files. Example:

• snd.yml:

general: "snd"
mgs: "snd/mgs"
pokemon: "snd/pokemon"

• snd/mgs/snd_pack.yml

/alert/: ["mgs_alert.mp3"]
/codec/: ["mgs_codec.mp3"]

Output of tree --dirsfirst:

.
├── snd
│   ├── mgs
│   │   ├── mgs_alert.mp3
│   │   ├── mgs_codec.mp3
│   │   └── snd_pack.yml
│   ├── pokemon
│   │   ├── battle.mp3
│   │   └── snd_pack.yml
│   ├── deja_vu.mp3
│   └── snd_pack.yml
└── snd.yml

By default, the bot will look for the snd.yml file in its directory. This can be customized; more on this in the "Setting up your own bot" section.

snd.yml syntax

Each line of the snd.yml file has to contain the name of the sound collection and a path to the directory with a snd_pack.yml file and the sound files. Please note there's no trailing slash.

snd_pack.yml syntax

Each line of the snd_pack.yml file has to have the keyword(s) specified as a regular expression and an array of sound(s) for that/these keyword(s). Look below for a Regex crash course.

Regex crash course

Some typical use cases are listed below, with explanations.

• /lol/i: ["laugh1.mp3", "laugh2.mp3"] This will play either laugh1.mp3 or laugh2.mp3 (chosen randomly). Note the i at the end - it makes the keyword case insensitive. This means that the sound will be played whenever someone posts a word or an emoji that contains lol, LOL, lOL, loL, LoL and so on.
• /fail|sad/: ["sad_trombone.mp3"] This will play sad_trombone.mp3 whenever someone posts a word or an emoji that contains fail or sad (so also failure and sadness).
• /kappa/: ["kappa.mp3"] This will play kappa.mp3 for any occurence of kappa. It will also work for a server emoji named kappa.
• /\:kappa\:/: ["kappa.mp3"] The server emoji :kappa: (and nothing else) will now play out kappa.mp3.
• /😂/: ["heavy_laugh.mp3"] This will play heavy_laugh.mp3 for the infamous laughing crying emoticon. You can add sounds to the standard Discord/Unicode emoticons this way.

Controlling the bot

The bot has a couple of commands. You need a "Manage Channels" permissions (or higher) to use them. The default prefix is "!".

• !soundmoji connect

Makes the bot connect to a voice channel you're in. You have to be in a voice channel in order to use this command.

• !soundmoji disconnect

Makes the bot disconnect from a voice channel.

• !soundmoji set_channels <channel1> <channel2> <...>

By default the bot listens for the keywords on all channels. You can, however, specify certain channel(s) for the bot to listen to. Example: !soundmoji set_channels general will make the bot listen to the messages on the #general channel. !soundmoji set_channels general soundmoji will make the bot listen to the messages on both the #general and #soundmoji channels.

• !soundmoji toggle_blocking

By default an user can change a sound which is already playing to another sound by posting a message with a keyword. You can toggle this behavior (if you do, keywords won't be accepted until the sound that was played out has finished playing).

• !soundmoji set_rate_limit_delay <new_rate_limit_in_seconds>

By default the bot will permit playing a sound every 3 seconds per channel. You can change this delay with this command. Example: !soundmoji set_rate_limit_delay 10 means that each channel will have to wait 10 seconds between playing a sound.

Help and sound listing

Some help is built in, you can trigger it by typing !soundmoji help.

You can also get the listing of the loaded sounds by typing !soundmoji list_sounds. Be wary that some servers may have lots of sounds, so you may get a lot of messages. It's usually better to see the sound listing on the server.

Setting up your own bot

Requirements

You require Ruby, libsodium, libopus and a couple of gems installed on your system. You also need to have a Discord Bot token and ID, see here for tips. You will need to enter them as environment variables SOUNDMOJI_BOT_ID and SOUNDMOJI_BOT_TOKEN, as appropriate.

Typical usage:

export SOUNDMOJI_BOT_ID=<discord_bot_id>
export SOUNDMOJI_BOT_TOKEN=<discord_bot_token>

After setting the appropriate variables, run the bot with ruby bot.rb.

Setting a custom soundpack path

By default, the bot will look for the snd.yml file in its directory. You can make the bot use another directory, though, by specifying the path as an argument.

Example:

ruby bot.rb /home/my_awesome_sounds

The bot will now enter the /home/my_awesome_sounds directory, look for the snd.yml file, parse it and start running.

You can also set your custom directory by setting the SOUNDMOJI_SOUNDS_PATH environment variable, like so:

export SOUNDSMOJI_SOUNDS_PATH="/home/my_awesome_sounds"

Running with Docker

You can run the bot with Docker Compose.

Example docker-compose.yml file:

version: '2'
services:
  bot:
    image: pgorni/discord-soundmoji:latest
    volumes:
    # You definitely want to change this. 
    - /some/path/on/host/discord-soundmoji-sounds:/snd
    environment:
      - SOUNDMOJI_BOT_ID=
      - SOUNDMOJI_BOT_TOKEN=
      - SOUNDMOJI_SOUNDS_PATH=/snd
    restart: unless-stopped

Edit it as you wish, save it as docker-compose.yml and then just run docker-compose up. (You have to have Docker Compose installed for this, of course.)

Donations

This software will always be free. Your help will be greatly appreciated! 💞

Instructions

1. Clone this repository.
2. Go into the bot's directory.
3. Run bundle install
4. Create the directories with sounds as explained before. You can create them in the bot's directory or somewhere else. In the latter case, you'll need to pass the location of the directory structure as an argument.
5. Run the bot with ruby bot.rb. If your sound files are somewhere else, you need to pass the location of the snd.yml file and the subsequent directories as an argument, for example ruby bot.rb /home/my-discord-sounds.
6. Invite the bot to your server with the link it displays.

License

GNU GPLv3.