openAI

Go to openai.com and signup for you API-key.

All chatbots are currently set to use the gpt-3.5-turbo engine

def generate_text(prompt):
    data = {
        "model": "gpt-3.5-turbo",
        "messages": [
            {"role": "system", "content": "You are an expert lawyer, and great at writing legal documents"},
            {"role": "user", "content": prompt}
        ]
    }

You can see the different text models at: https://platform.openai.com/docs/models/gpt-3

Chatbots using the speech_recognition framework are set to the google speech API. You can use openAI Whisper if you prefer(it runs offline), read the docs at: https://pypi.org/project/SpeechRecognition/.

You need the Pyaudio framework to capture audio from microphone. If you're running macOS, you might run into an error using pip install pyaudio. Install portaudio before pyaudio.

brew install portaudio
pip install pyaudio

I don't know why but it works.

Dependencies

pip install openai
pip install SpeechRecognition
pip install requests

Have fun!