Machine Learning-based Tamil Sign Language Recognition Tool for the Deaf and Hard-of-Hearing Community.

This paper proposes a machine learning-based Tamil sign language recognition tool for the deaf and hard-of-hearing community. The tool uses computer vision techniques to capture the hand gestures of sign language users and processes them using convolutional neural networks (CNN) to recognize the corresponding Tamil alphabet or word. The proposed system consists of two main stages: data collection and model training. During the data collection stage, a dataset of Tamil sign language gestures is created by recording videos of sign language users performing various gestures. In the model training stage, the dataset is preprocessed and augmented, and a CNN is trained to recognize the gestures. The proposed system achieves a recognition accuracy of 85% on the test dataset, which is a significant improvement over the existing system. The proposed tool can be deployed on a mobile device or a web application, making it easily accessible to the deaf and hard-of-hearing community.