ayushpai

1st in open weight

Conversational AI with GPT-4 Vision, OpenAI Whisper, and TTS.

This project integrates GPT-4 Vision, OpenAI Whisper, and OpenAI Text-to-Speech (TTS) to create an interactive AI system for conversations. It combines visual and audio inputs for a seamless user experience.

I created version 2.0 of my AI Basketball Referee. I trained a custom YOLOv8 deep learning model with over 3000 images. The system can accurately detect travels and double dribbles. I’ll be expanding it soon to other basketball violations.

I would love any feedback to make this even better!