People who are blind or partially sighted rely significantly on their other senses, it being the sense of touch or auditory senses, to comprehend their surroundings. It is extremely difficult for a blind person to determine the objects that might come in their path without coming in contact with it. When travelling from one location to another, a blind person finds it difficult to notice the presence of any obstacles in their path. Physical touch between a person and an object can be hazardous, if not deadly. This research aims to convert the visual world into an aural world capable of informing blind individuals about items and their spatial placements. Objects recognized in the scene are identified and then transformed to speech. On the client side, video is recorded with a camera device and sent to the server for real-time image identification. We will be using different algorithms like YOLO, Darkflow (YOLO and TensorFlow), Darknet.