This is a very simple demo of a speech-to-text and then text-to-speech AI.
To get this running, follow the below steps:
- Clone this repo:
git clone https://github.com/nydasco/jen-ai.git
- Within the folder, create a virtual environment:
python3 -m venv .venv
- Start the environment:
source .venv/bin/activate
- Install dependencies:
pip3 install -r requirements.txt
- Download the LLM (5.13G):
huggingface-cli download TheBloke/OpenHermes-2.5-Mistral-7B-GGUF openhermes-2.5-mistral-7b.Q5_K_M.gguf --local-dir . --local-dir-use-symlinks False
- Run the file:
python3 jen.py
Note that the first time it runs it will take a while to get started. There are additional models that need to be downloaded. They're smaller than the main LLM.