Vision loss may be a drawback that affects individuals all around the world. As pcs have gotten higher at understanding pictures because of advances in computer vision, the idea of a virtual assistant for the blind that would scan text1, identify/spot objects2, or maybe describe an entire scene in natural language3 is turning into more and more realistic. In this project, I created an Associate in Nursing robot application capable of reading aloud Network, text (e.g. product labels, price tags, text on clothing). the appliance uses a classifier trained exploitation the MS-COCO4 and COCO-Text5 datasets to seem for text and uses Google Cloud Vision to extract the text.