GLaDOS Personality Core

This is a project dedicated to building a real-life version of GLaDOS.

This is a hardware and software project that will create an aware, interactive, and embodied GLaDOS.

This will entail:

• Train GLaDOS voice generator
• Generate a prompt that leads to a realistic "Personality Core"
• Generate a MemGPT medium- and long-term memory for GLaDOS
• Give GLaDOS vision via LLaVA
• Create 3D-printable parts
• Design the animatronics system

Software Architecture

The initial goals are to develop a low-latency platform, where GLaDOS can respond to voice interactions within 600ms.

To do this, the system constantly records data to a circular buffer, waiting for voice to be detected. When it's determined that the voice has stopped (including detection of normal pauses), it will be transcribed quickly. This is then passed to streaming local Large Language Model, where the streamed text is broken by sentence, and passed to a text-to-speech system. This means further sentences can be generated while the current is playing, reducing latency substantially.

Subgoals

• The other aim of the project is to minimize dependencies, so this can run on constrained hardware. That means no PyTorch or other large packages.
• As I want to fully understand the system, I have removed a large amount of redirection: which means extracting and rewriting code. i.e. as GLaDOS only speaks English, I have rewritten the wrapper around espeak and the entire Text-to-Speech subsystem is about 500 LOC and has only 3 dependencies: numpy, onnxruntime, and sounddevice.

Hardware System

This will be based on servo- and stepper-motors. 3D printable STL will be provided to create GlaDOS's body, and she will be given a set of animations to express herself. The vision system will allow her to track and turn toward people and things of interest.

Installation Instruction

New Simplified Windows Installation Process

Don't want to compile anything? Try this simplified process, but be aware it's still in the experimental stage!

1. Open the Microsoft Store, search for python and install Python 3.12.
2. Download and unzip this repository somewhere in your home folder.
3. Run the install_windows.bat. During the process, you will be prompted to install eSpeak-ng, which is necessary for GLaDOS's speech capabilities. This step also downloads the Whisper voice recognition model and the Llama-3 8B model.
4. Once this is all done, you can initiate GLaDOS with the start_windows.bat script.

Regular installation

If you want to install the TTS Engine on your machine, please follow the steps below. This has only been tested on Linux, but I think it will work on Windows with small tweaks. If you are on Windows, I would recommend WSL with an Ubuntu image. Proper Windows and Mac support is in development.

1. Install the espeak synthesizer according to the installation instructions for your operating system.
2. Install the required Python packages, e.g., by running pip install -r requirements.txt on Mac or Linux systems without an Nvidia GPU, and pip install -r requirements_cuda.txt if you have a modern Nvidia GPU.
3. Download the models:
   1. voice recognition model
   2. Llama-3 8B or
   3. Llama-3 70B and put them in the ".models" directory.
4. For voice recognition, we use Whisper.cpp
   1. You can either download the compiled whisper.cpp DLLs (recommended for Windows), and copy the dll to the ./submodules/whisper.cpp directory
   2. Or compile them yourself.
      1. To pull the code, from the GLaDOS directory use: git submodule update --init --recursive
      2. Move to the right subdirectory: cd submodules/whisper.cpp
      3. Compile for your system (see the Documentation), e.g.
         1. Linux with CUDA: WHISPER_CUDA=1 make libwhisper.so -j
         2. Mac with CoreML: WHISPER_COREML=1 make -j
5. For the LLM, you have two option:
   1. Compile llama.cpp:
      1. Use: git submodule update --init --recursive to pull the llama.cpp repo
      2. Move the the right subdirectory: cd submodules/llama.cpp
      3. Compile llama.cpp, (see the Documentation)
         1. Linux with CUDA make server LLAMA_CUDA=1
         2. MacOS with Metal make
   2. Use a commercial API or install an inference backend yourself, such as Ollama or Llamafile:
      1. Find and install a backend with an OpenAI compatible API (most of them)
      2. Edit the glados_config.yaml
         1. update completion_url to the URL of your local server
         2. for commercial APIs, add the api_key
         3. remove the LlamaServer configurations (make them null)

Help Section

1. If you have an error about packages or files not being found, make sure you have the whisper and llama binaries in the respective submodules folders! They are empy by default, and you manually have to add the binaries as described above!

2. Make sure you are using the right Llama-3 Model! I have made Llama-3 8B, with the quantization Q6_K the default. You might need to redownload the model if you don't have Meta-Llama-3-8B-Instruct-Q6_K.gguf in your models folder!

3. If you have limited VRAM, you can save 3Gb by using downloading a highly quantised IQ3_XS model and moving it to the models folder. If you do this, modifiy the glados_config.yaml to modify the model used: model_path: "./models/Meta-Llama-3-8B-Instruct-IQ3_XS.gguf"

4. If you find you are getting stuck in loops, as GLaDOS is hearing herself speak, you have two options:

   1. Solve this by upgrading your hardware. You need to you either headphone, so she can't physically hear herself speak, or a conference-style room microphone/speaker. These have hardware sound cancellation, and prevent these loops.
   2. Disable voice interruption. This means neither you nor GLaDOS can interrupt when GLaDOS is speaking. To accomplish this, edit the glados_config.yaml, and change interruptible: to false.

Windows Run

Prerequisite WSL2 with fresh drivers, here is guide https://docs.docker.com/desktop/gpu/

1. git submodule update --init --recursive
2. put models in models dir or mount that dir into a docker container
3. docker build -t glados .
4. docker run -e "PULSE_SERVER=/mnt/wslg/PulseServer" -v "/mnt/wslg/:/mnt/wslg/" --gpus=all --ipc=host --ulimit memlock=-1 --ulimit stack=67108864 glados

It works in ubuntu terminal started with WSL2

Running GLaDOS

To start GLaDOS, use: python glados.py

You can stop with "Ctrl-c".

Testing

You can test the systems by exploring the 'demo.ipynb'.

Star History