This paper presents a machine learning-based framework aimed at enhancing mobility and spatial awareness for the visually impaired by reviewing different papers where authors suggest various object detection models, particularly focusing on different versions of the YOLO algorithm, to provide real-time analysis of a user's surroundings. The reviewed systems are audio feedback integrated, converting visual data into descriptive audio cues to help users identify objects and obstacles, enabling independent navigation. The paper also reviews the use of the Darknet framework with TensorFlow, and tools such as Google Text-to-Speech (GTTS) and Raspberry Pi, emphasising their role in building effective, real-time assistive technologies.