Blind individuals face significant challenges in navigating their surroundings independently. To address this, we propose an intelligent Blind Assistance System that integrates computer vision, machine learning, and IoT to provide real-time object recognition and environmental awareness through audio feedback. The system consists of a wearable camera that captures live video, an ML model that performs object detection and scene recognition, and an NLP-based voice assistant that converts detected objects into speech output. Additionally, an Arduino-based ultrasonic sensor detects nearby obstacles and triggers a buzzer for proximity alerts. The system is further enhanced by a Flutter-based mobile application, which utilizes a TensorFlow Lite (TFLite) MobileNet SSD model for live object detection and offers voice-controlled interactions. To improve accessibility, a Flask/FastAPI server hosts the scene detection model, allowing seamless integration into the mobile app. This low-cost, AI-powered assistive solution empowers visually impaired individuals by enhancing their spatial awareness, improving mobility, and promoting independent living.