EchosTeam4

Objective:

Create a Bot that monitors a Voice Channel and utilizes a speech-to-text engine/service to caption voice chat for Deaf & Hard of Hearing. May be able to utilize Discord’s overlay as the surface to display captions on.

Tools:

NPM + Node.js
Electron
Socket.io
FFMPEG
Jquery
Google Speech To Text

Requirements:

Google Cloud Account
Discord Account

How to use:

Turn bot services on
Join voice channel
Type <prefix>activate <channel>
Clients can use Electron
Speak

How to Configure:

You need a config.json with the following:

{
    "prefix": "...",
    "outputSocketPort": "...",
    "token": "..."
}

Testing

The bot was tested as much as possible to get the best user experience. The bot was able to recognize most voices.

Future Work:

Work on improving speech recognition
Use machine learning for better text results
Improve overlay design

cshadd/EchosTeam4