Published Paper Details:

The last decade saw rapid advancements in different fields, and with the advancements of deep learning technol- ogy,remarkable progress has been made.That said, obstacles still remain with regards to Sign Language(SL) recognition, SL translation, and SL generation. Models that are used to assist people with hearing impairments and SL users are still not accurate or visually appealing. In this study, we present creative strategies that will establish the essential elements of a complete system to encode, decode and instantaneously interpret SL. As a way of dealing with a two-dimensional plane rotation for better recognition accuracy and continuous gesture recognition, we utilize the MediaPipe library with a hybrid model driven by convolutional neural networks (CNNs) and bi-directional long short term memory (Bi-LSTM) for pose extraction and text synthesis.In addition, multilingual translations are made possible by including Google Translate API in the system.