Paper currency recognition is a significant challenge for visually impaired individuals, especially in financial transactions. To address this, a real-time mobile application is presented, enabling users to identify Libyan banknotes quickly and accurately without external hardware or an internet connection. This approach integrates YOLOv11, a state-of-the-art deep learning model, into native Android (Kotlin) and iOS (Swift) applications. The model is trained on a diverse dataset of Libyan banknotes under various conditions to ensure robust detection. Once a banknote is identified, the system provides instant auditory feedback via Text-to-Speech (TTS) technology. The model achieves a high precision of 99.23%, recall of 99.05%, and mAP50-95 of 99.11%, demonstrating strong detection performance. Extensive testing demonstrates that our YOLOv11-based system achieves high accuracy and real-time performance, making it a practical and efficient tool for assisting visually impaired individuals in financial transactions. By implementing this advanced object detection model directly within mobile applications, our solution enhances accessibility and promotes financial independence.