In this presentation, we will be demonstrating a Computer Vision demo using YOLOv5 on the American Sign Language Dataset including 26 classes.The model identifies signs in real time as well as with input image or audio and builds bounding boxes showing label with confidence value..The model is showcased using streamlit which can take input as an image.