Timmy is designed to be an avatar assistant, supporting Speech Recognition, Text to Speech, LLM capabilities, and of course, an avatar.
You can clone the repository using git clone https://github.com/IsaiahSama/Timmy to your device! You'll also need Python 3.10 or higher for the program to run properly.
You can then install the libraries by running pip install -r requirements.txt. I highly recommend modifying the config.yaml file, primarily the ROLES section, and changing the prompt for the GENERAL one (or making your own and updating the ROLE entry to match). With this, the last thing you'll need is an API key for either GEMINI, or Open AI, and placing that key in a .env file based off of the .env.sample file. Be sure to update the MODEL entry of the config.yaml file to reflect the model you wish to use.
From here, you can run python main.py or python3 main.py to get started. Have fun!
I recently created a post on Medium that goes into my thought process and implementation when creating this, so feel free to give that a read and some claps as well if you enjoy 😄. Timmy The AI Assistant.