https://github.com/vagr9k/voicebot

https://github.com/vagr9k/voicebot

discord discord-bot discord-js hotword-detection multithreading n-api nodejs speech-recognition speech-to-text typescript

Last synced: about 2 months ago
JSON representation

• Host: GitHub
• URL: https://github.com/vagr9k/voicebot
• Owner: Vagr9K
• License: agpl-3.0
• Created: 2019-03-02T20:01:01.000Z (over 7 years ago)
• Default Branch: master
• Last Pushed: 2022-06-22T16:41:56.000Z (about 4 years ago)
• Last Synced: 2025-11-09T05:03:26.846Z (8 months ago)
• Topics: discord, discord-bot, discord-js, hotword-detection, multithreading, n-api, nodejs, speech-recognition, speech-to-text, typescript
• Language: TypeScript
• Size: 319 KB
• Stars: 0
• Watchers: 1
• Forks: 1
• Open Issues: 25
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# VoiceBot

VoiceBot is a discord bot that support voice commands.

Invite the bot to your channel and then pronounce `{keyword} play {song name}` to play your favorite song!

## Demo

[![Demo](https://img.youtube.com/vi/vRwp--RoJdo/0.jpg)](https://www.youtube.com/watch?v=vRwp--RoJdo)

Why `Terminator`? It seems to be the most consistently detected hotword.

Why `Mozart`? It's not copyrighted.

## How does this work

Consult [my blog post](https://vagr9k.me/using-n-api-for-high-performance-voice-command-detection-in-discord) about the design decisions if you're interested.

## How to use

NOTE: This project is in an alpha stage an relies on a native N-API module [native-voice-command-detector](https://github.com/Vagr9K/native-voice-command-detector) for high performance, multithreaded hotword detection and speech recognition. Consult [its documentation](https://github.com/Vagr9K/native-voice-command-detector#building) on how to build it.

To use the bot:

- Setup a `.env` configuration file:

  ```

    DISCORD_BOT_TOKEN=YOUR_BOT_TOKEN

    DISCORD_BOT_PREFIX='terminator ' # Or any other prefix used for text commands

    YOUTUBE_API_KEY=YOUR_YOUTUBE_API_KEY # For searching songs

    GCLOUD_SPEECH_TO_TEXT_API_KEY=YOUR_CLOUD_KEY # For native-voice-command-detector

    PV_KEYWORD_PATH= # For native-voice-command-detector

    PV_MODEL_PATH= # For native-voice-command-detector

    PV_SENSITIVITY=0.5 # For native-voice-command-detector

    MAX_VOICE_BUFFER_TTL=200 # For native-voice-command-detector

    MAX_COMMAND_LENGTH=3000 # For native-voice-command-detector

    MAX_COMMAND_SILENCE_LENGTH=1000 # For native-voice-command-detector

  ```

  Consult [native-voice-command-detector](https://github.com/Vagr9K/native-voice-command-detector) documentation for the explanation of the last parameters.

- Install the dependencies via `yarn`

- Launch the bot via `yarn start`

## Copyright

Copyright (c) 2019 Ruben Harutyunyan