dnhkng GlaDOS - Your Own, Free & Unhinged AI Companion!
HTML-код
- Опубликовано: 27 май 2024
- Yes. There's nothing new about speaking to an AI and having it speak back, however, when the responses are this fast then things start to become a bit more fun - especially with an AI as unhinged as this one...
dnhkng GlaDOS is free, open-source goodness, meaning you can download and run on your own computer in the comfort of your own home. Everything runs locally, so no API keys to mess around with either. Yay!
NB. This is not a valve product, and yes - they probably need a better name for the project ;)
Want to support the channel?
/ nerdyrodent
Links:
github.com/dnhkng/GlaDOS
== More Stable Diffusion Stuff! ==
* Installing Anaconda for MS Windows Beginners - • Anaconda - Python Inst...
* Installing ComfyUI for Beginners - • How to Install ComfyUI...
* ComfyUI Workflows for Beginners - • ComfyUI Workflow Creat...
* Faster Stable Diffusions with the LCM LoRA - • LCM LoRA = Speedy Stab...
* Make an Animated, Talking Avatar - • Create your own animat...
* Make A Consistent Character in ANY pose - • Reposer = Consistent S...
== Contents ==
0:00 - dnhkng GlaDOS Intro
4:43 - dnhkng GlaDOS Install - Наука
Now to make it interrupt the user. I wonder how one would implement that.
Soon, hopefully!
Unhinged sounds nice
Oh boy I'm about to get Alexa very mad at me...
This feels like a personal attack on Shodan.
😉
this is awesome. I got so hooked on AI stuff I started working on my own one. Possibilities are endless.
Who needs that when I have you? 😁
😉
The system shock community is ECSTATIC real now.
thanks ❤, can you make video for stable-makeup for windows11 ?
My nerdy friend 🤘😉💕
thx for the vid
*Quick question Mr NerdyRodent, is your bash script for windows Saaar ????*
The Linux bash script is for Linux, but expect some MS Windows support in the coming months! If you’re going the whole way though, for the full robotic setup you’ll want to use Linux.
@@NerdyRodent I am saving up to get larger storage so I can dual boot.
There is now a windows branch in the repository with a bash script that does everything for you!
Will this work on Windows in the same way? I was having trouble compiling whisper so was wondering if your compiled bash script would work :)
Overall, it's easier to just use Linux as then you're done in less than 15 minutes - but yes, whisper will compile even on Windows. Just need a bit of tweaking in glados for Windows weirdness, and you're done :) Probably best to wait a few weeks for Windows people to do their thing... unless you like the adventure!
@@NerdyRodent I tried in WSL2 but couldn't get microphone access :( Followed a few dead-ends involving weird streaming audio solutions but found a github conversation in microsoft/wsl which states that (as of 3 days ago) there still isn't any official solution
Was considering signing to the patreon for the bash script, but from what you say, sounds like it won't fix all the issues with a Windows installation.
@@NerdyRodent Thank you, yeah so excited to try this but will be patient for windows as my WSL environment is shot for some reason). Also appreciate and love all your work :) And as a rodent lover you might like to know I will be getting 2 pet rats in a couple weeks, excited to add the Rodent to my Nerdy haha.
Yup, overall, it’s just easier to use Linux!
Yay! Rats are awesome :) Personally I'd avoid WSL and just go for a normal Linux install, simply because WSL brings it's own set of Windows nightmares, but each to their own. There have been a couple of updates already, so let's see what the coming weeks bring!
Neato 😮
Hey Nerdy, how is it not hearing its own voice for the interrupting aspect... do you have the output audio isolated? Seems like some voodoo going on here.
Yes! I am using some advanced, alien technology called… headphones! 😉
@@NerdyRodent Thats a bummer.. so if you were to use a set of speakers it won't work because it will hear its own voice? The issue I had when making my voice assistant was that you had to stop listening when the Ai was speaking, and then when it was done, listen until the person stops speaking. I love the barge in, it would be amazing if it had some kind of very advanced echo cancellation so it didnt hear its own voice.
@@RchGrav The echo cancellation is not that advanced, as you need a measurement of the audio roundtrip latency between when a frequency is sent out by the program, and when it is received.
Then it would be a matter of mixing the received audio signal - with the original "output speech", except with its polarity flipped and delayed by this roundtrip latency.
Existing impulse response ("IR") tools basically handle the first part (they even feature the expected frequency distortion of the signal by the speaker+microphone, how neat!), the latter signal processing part of "record input of Time Period, apply measured IR to cached synthesized speech for Time Period, flip polarity" is simple and straightforward.
And voila, it is as if the robot speech was all in your head! To its ear, anyway.
only a few days until I become an Arkham horror... ;)
also is this a good idea to install? Like... uhh.. really? Why isn't anyone trying to invent the portal gun? come on people!
Oh. It’s totally fine. Wait. Why are my lights flickering? What the….
@@NerdyRodent uh oh.
I wanna make real Claptrap now.
😉
No fun for plain Windows users, ha?
It even runs on Windows too, yes!
*Is your bash script for windows Mr NerdyRodent Saaar?*
Do you plan to make a tutorial but on Windows?
I'm "stealing" the code for interruptions
Does it work for speech in other languages?
Can you do a HiDiffusion ComfyUI tutorial?🙏
Ah, you’re more a diffusers pipeline kinda person, eh? 😀