About 2.4 billion people worldwide are affected by vision impairment, which has a major effect on their ability to identify currencies and recognize objects. We propose Vision-Mate, a low-cost and user-friendly Android application developed and implemented to assist visually impaired people, as a solution to these challenges. For object detection, the suggested solution uses a pre-trained deep learning model, MobileNet SSD, which was trained on the COCO data set and obtained an accuracy rate of 89%. Furthermore, for currency detection, the MobileNetv2 technique is trained using the Indian currency data set, which shows the training and validation accuracies 83. 79% and 85. 37%, respectively. Furthermore, our CNN model achieved a 100% recall rate for classifying 200-rupee notes and a 97% precision rate for the classification of 20-rupee notes. Similarly, a transfer learning model implemented using MobileNet SSD recorded a 96% precision rate for the 20-note currency and a 97% recall rate for the 500-note currency. Furthermore, ‘VisionMate’ is enhanced with voice-assisted activity commands, making it easier to use and more accessible for people with vision impairment. This work contributes a practical and innovative tool with diverse features designed to help and enhance the lives of visually impaired people.