https://github.com/dnhkng/GlaDOS
"This is the Personality Core for GLaDOS, the first steps towards a real-life implementation of the AI from the Portal series by Valve."
The main thing about this one, is how fast it can generate responses :) You can actually have a (nearly) normal conversation!
The attached installation bash script is designed to be run on a standard, Linux+Nvidia system in whatever conda environment you've created first (such as with):
conda create --name glados python=3.11
conda activate glados
You can then run
bash GlaDOS.sh
For other operating systems / GPUs, follow the instructions for each component that relate to your setup:
1. eSpeak-ng - https://github.com/espeak-ng/espeak-ng/blob/master/docs/guide.md
2. Llama.cpp - https://github.com/ggerganov/llama.cpp
3. Whisper.cpp - https://github.com/ggerganov/whisper.cpp
Note that the project is in it's very early stages, so if progress continues then things are likely to change considerably in the future!