Advancing Human-Computer Interaction: Real-Time Gesture Recognition and Language Generation Using CNN-LSTM Networks

Efficient human-computer interaction (HCI) has pushed forward hand gesture recognition (HGR), and natural language utterance generation (NLUG). Traditional systems are limited because these rely on static infrastructure and manual processing, accompanied by inaccuracies and latency. In comparison, the proposed system uses an advanced fusion of convolutional neural networks (CNNs) for gesture recognition and long short-term memory (LSTM) networks for language generation to achieve real-time processing with context-aware responses. The results of the experiments showed a 90.5% increase in precision for gesture recognition and BLEU score 0,715 NLUG were described as displaying superiority to existing systems on the tool performance The proposed approach also reduces the latency to 90 ms, while in traditional systems it is about 150 ms. These results suggest an opportunity to reimagine HCI systems as truly immersive and efficient, responsive experiences that can optimize not only user interactions with applications in specific domains, but also social computing at large.