Skip to content

Latest commit

 

History

History
9 lines (6 loc) · 545 Bytes

README.md

File metadata and controls

9 lines (6 loc) · 545 Bytes

GPT-3_based_AI

A simple program that gives a voice to open-ai's gpt-3 model.

The program uses speech recognition module (with google speech recognition) to recognize your voice, and passes reading to Open-AI's API, the reply the API gives is fed into the another module called pyttsx3 which says the reply out loud.

It uses Open-AI's davinci model, though this can be modified in the code.

You will also have to sign up to Open-AI, to generate an API key, and then create an Environment variable called OPENAI_API_KEY with your API key.