Voice Controlled Discord Bot

This repository is for the 5th Hackathon of Documatic. With that Project I am aiming to place #1 and win 50€

About

It is a voice controlled Discord Bot, means you (mainly) do not use prefix / Slash Commands, no! You will be using your VOICE.

Ofc. you somehow need to parse that audio, for that I am using https://wit.ai which is a free speech-to-text ai, which you need to train! That means it might sometimes not return what you say... but I tested it over 19 months on another Project, and there it works 100% everytime, even multilangual now..

It is a Music Bot, as other interactions are not that optimal, like kick / ban, you can't control that with voice 100% sure...

Check out the Show-off and Tutorial Video

System-Requirements

• Idle Load: 17mb Ram
• High Load: 25-30mb Ram
• Highest Memory Load with HEAPS for 2 Days runtime: 97mb Ram
• JS-Engine: nodejs v16.10 or later
• RUST (cmake): latest
• FFMPEG: npm i ffmpeg-static ffmpeg & apt-get install -y ffmpeg

Example Usage

How to use the Bot? | Check the Video

0. Join the testing Server - Prefix: v! / self-host it!
1. Join a Discord Voice-Channel, in a Server, where the Bot is in!
2. Type in a Text-Channel v!control or @Bot control a. Now it'll only listen and be controlable by YOU
3. Speak your COMMANDS and add a KEYWORD in beforehand. a. Examples: i. bot play shape of you ii. bot play thunder iii. bot play despacito iv. bot nightcore =Audio-Filter v. bot skip vi. bot stop
4. Tipps to get understood more often! a. Make sure to talk in a normal speed and loudness, do not scream or errape b. Also Make sure to Reduce background noices, speak clear and fluently to be "recogniced" by the bot pretty well!

Self-Hosting

1. Download the repo | Click-here
   • or: git clone https://github.com/Tomato6966/voice-controlled-discord-bot
2. Rename example.env to .env and fill out the variables (get wit.ai Server Access Token from wit.ai)
3. yarn install / npm install (make sure you have rust & nodejs on your system)
4. Type node index.js / npm start

Want to add more music commands?

Take a look at my Light-Music-Bot Project, which is similar to this one (for the music system) https://github.com/Tomato6966/light-music-bot

Resources used (modules & credits)

• node-fetch@2 for api Calls
• discord.js@latest as my Discord-Bot-Wrapper
• @discordjs/voice, @discordjs/opus, discord-ytdl-core, ytdl-core, youtube-sr, ffmpeg, libsodium-wrappers for the Music System (Its similar to my light-music-bot)
• ffmpeg, prism-media, node-crc for parsing / Piping / Transforming Audio Streams, -Buffers and -Files.
• dotenv for allowing to use .env ENVIRONMENT Variables

Explanation and show off Video

ToDo

• ✅ (28/07/2022 20:17) Change Message Responses from the voice-command, to VOICE-RESPONSES (Bot speaks it)
  • Announce Songs: "Now playing XYZ"
  • also for Skip: "Skipping the current Track"
• ✅ (27/07/2022 18:23) Make the structure better, and fix all bugs
• ✅ (27/07/2022 18:23) Add all to a .env file, and create an example.env file
• ✅ (27/07/2022 18:23) Adjust the Settings
• ✅ (27/07/2022 18:23) Make it multi langual without any modules
• ✅ (27/07/2022 18:23) Try to make it even faster ;)