Indonesian Sign Language (BISINDO) Alphabet Detection Using the You Only Look Once (YOLO) Algorithm Version 8

People with disabilities have increased over the past few years in Indonesia. Data from the Central Bureau of Statistics, Republic of Indonesia shows that the number of people with disabilities reached 212,240 in 2022 and specifically deaf people reached 19,392. This shows the need for attention to people with disabilities, especially the deaf and hard of hearing. This research aims to develop an Indonesian Sign Language (BISINDO) alphabet detection model using the You Only Look Once (YOLO) v8 algorithm. The resulting model is expected to help people in understanding sign language, thus expanding accessibility for people with disabilities. The model was trained using 2.759 images and 30 videos. The results show that the model can identify 26 letters in BISINDO alphabet including letters with motion (G, J and R) with an accuracy rate of 99%, a precision value of 99.4%, recall 98.4% and F1-score reached 98.9%. These values indicated that the model works well in predicting the BISINDO alphabet. The model built using this YOLOv8 produces sign language gesture conversion output into text, which can then be assembled into words, thus facilitating communication with deaf and hard of hearing people.