This is a project by protogenraymond that enables some kind of voice detection in your games. It'll repeat what it believes you said in chat for you.
You should not use this in large games. This is merely an educational project, so use at your own risk!
This also won’t work without using Chrome. I plan on getting rid of that requirement soon, but I haven’t done any research on other voice detection apis. Sorry
Currently requires - Google Chrome - Node.JS - Access to your router’s port forwarding menu*
All the source is now on the GitHub repository! It’ll now always be there so you can easily view the source.
This link is dead as the repository and this Github account was deleted, but I will keep it on here for archival reasons.
As previously mentioned in the last update, this concept broke TOS by inherently redirecting users off-platform. This was NOT voice chat, but rather speech-to-text. It used one of Google’s Web APIs and it wasn’t efficient nor reliable as a form of messaging. It was merely a concept that I was toying around with but would’ve been better had I not released it. Roblox’s native voice chat is significantly better and is supported in games without the need for off-site redirections.
TL;DR, Don’t use this, just use Roblox’s native voicechat solution.
For 1.1, I plan on having a much better UI (that I’m making with Adobe XD currently), instead of an ugly red button and a textbox. It may also include a way of verifying your user in-game!
It’s a mash between a website and script. You have to use Google Chrome for recording, Node.js handles the voices sent from Google Chrome, and the script takes it from there to make the player chat what the browser thinks the player said.
It is filtered, by the way, so it can’t break Roblox rules unless forced to do so. This breaks Roblox’s TOS by redirecting users off-site. Do not use this (if you still have access to it’s original source).
Seems like a cool and somewhat unique idea, but when people have done this stuff before (those hackers and those voice chat games) it never went out well. You can never 100% censor someones voice using a code, and I believe if you tried it’d really limit what voice chat can do, constantly producing false positives if the filter is too strong with banned words.
Also it’s really hard to trust a project like this when no source code is given. Sure we can check it out in our games, but how do we know this is not a dangerous script posed as a perfectly working and reliable one? Are we using a random site we’ve never been to before? If not, is it a plugin for the Chrome Web Browser?
My original response was basically just “This is voice detection, not voice chat. It uses the Roblox filter” but with attitude. Sorry!
[ORIGINAL RESPONSE]
All the source is currently inside the zip; I’m going to upload it to GitHub for easier access. Also, this isn’t actual voice chat. It doesn’t repeat what you said as an audio, it just uses Google’s Speech Recognition API.
It’s not my fault if Roblox’s filter fails to censor something, I don’t control the filter.
If you modify the script to remove the filter, then that is all on you.
It’s giving be the “Failed to connect to server” error when I’ve entered my global ip as the value (as a string). Is there anything I can try to do to fix this?
Okay, this clears somethings up. I understand with the whole removing filter stuff and you’ll be responsible yada yada, but it’s more practical to upload to GitHub before releasing the project to the world. There’s always going to be people who care about that. (who wouldn’t?)
Make sure HTTP Requests are enabled, and that you haven’t closed out of the command prompt that has [SERVER]: Webapp is open at localhost:80. It won’t work un-published.
The way Roblox Studio works is that requests are made from your device, unlike the Roblox Client where it makes requests from the Roblox servers themselves, meaning you only need to port forward when attempting to play the game through the client, as the servers cannot access your device’s 80 port without it being pf-ed.
