The inability of visually impaired people to independently recognize and handle currency can have an adverse effect on their capacity to conduct everyday activities and conduct financial transactions. This research is implemented with the most efficient object identification algorithm YOLOv8 along with the Roboflow annotation tool for effective dataset labeling to construct an inclusive cash detection system and identify fake invalid currency for visually impaired people. The dataset, which consists of various pictures of counterfeit Indian currency notes, is annotated to produce bounding boxes around the currency regions. Following the trained YOLOv8 model’s picture prediction, the discovered currency information is processed using the GTTS (Google Text-to-Speech) library to generate speech. Through this integration, the system may speak out the details of the detected currency, including its type, denomination, and validity or falsity, giving visually impaired individuals access to real-time information. To ensure strong performance, the system’s evaluation takes into account important metrics like F1 score, precision, recall, and mean average precision (mAP). When combined with GTTS, the final model offers a complete solution that improves accessibility by offering auditory money identification for the person who is visually impaired. Therefore, cash detection technology plays a vital role in empowering visual impaired human being by giving them the means to handle their finances on their own, encouraging autonomy, and promoting inclusion in a various spheres of life.