An AI-based sign language translation system that converts sign language videos into meaningful text sentences. It uses CNN (ResNet18), BiLSTM, and an attention-based LSTM decoder to learn spatial and temporal gesture patterns, enabling automatic translation using the RWTH-PHOENIX-Weather 2014 dataset.