Smart Object Detection and Audio Feedback System for the Visually Impaired

This paper proposes a new assistive system, "Sight Through Sound," that is intended to improve the mobility and independence of visually impaired people. With real-time detection of objects and audio-based feedback, the system overcomes current solution shortcomings including high expenditure, inability to adapt in real time, and poor language support. The proposed system combines YOLO and Faster R-CNN-based advanced object detection algorithms with IoT-based hardware including Raspberry Pi and ESP32-CAM for seamless processing. Obstacles are detected using ultrasonic sensors and a kannada text-to-speech (TTS) module is used to deliver localized audio cues. Results from experiments show greater accuracy and responsiveness than existing technologies and provide a scalable and cost-effective solution that enables visually impaired users to move independently with greater situational awareness.