Sign Language Recognition and Training Module

Abstract

Abstract Recognition of sign language has long been essential to communication between the deaf and non-verbal cultures. From early electric signal-based sign language identification to more recent recognition using machine/deep learning techniques, researchers from all over the world have made an effort to automate this process. Key point detection-based sign language recognition (SLR) is the major goal of this study. The focus of this work is American Sign Language (ASL), particularly ASL pickle data. Several machines learning algorithms, including random forest, support vector machine, and k nearest neighbour, were used to train the model. Lastly, the best model is selected from the model testing using assessment parameters including f1score, precision, and recall. To gather user input, a simple GUI is made, and the forecast is then made using the best machine learning model. Additionally, a training tool for learning American sign language has been developed, which will have a significant impact on non-verbal cultures.