An Optimized Sign Language Recognition Using Convolutional Neural Networks (CNNs) and Tensor-Flow

Sign language is an essential means of communication for people with hearing disabilities. However, there is often a communication gap between hearing and non-hearing individuals, which can lead to social exclusion and discrimination. To bridge this gap, They proposed a deep learning-based sign language interpreter using TensorFlow. The system can recognize different signs and interpret them into text or speech output, allowing non-hearing individuals to communicate with hearing individuals more effectively. To evaluate the system on a publicly available sign language dataset and achieve an accuracy of 95% for recognizing signs and 90% for interpreting them into text or speech output. The results demonstrate the effectiveness of the approach and highlight the potential of deep learning for improving accessibility and inclusion for people with hearing disabilities. The main result of this research is to get the accuracy and accurate prediction of the model that they will use for transfer learning with the dataset