Protein secondary structure (PSS) prediction plays a pivotal role in explaining protein folding patterns and understanding protein function. In this study, we propose a novel approach for protein secondary structure prediction using Hybrid Convolutional Neural Network (CNN) architecture with an integrated Attention Mechanism. Leveraging the rich spatial information encoded in protein sequences, our model aims to accurately predict the secondary structure elements, including alpha helices, beta strands, and coils. The Hybrid CNN architecture combines convolutional layers for feature extraction with attention mechanisms for focusing on informative regions within the protein sequences. We evaluated the performance of our model on benchmark datasets, including CB513 and CASP12, using standard evaluation metrics such as Q8 accuracy and Q3 accuracy. The results demonstrate the efficacy of our approach in achieving state-of-the-art performance in protein secondary structure prediction, outperforming existing methods. Our study underscores the potential of deep learning techniques, specifically Hybrid CNN with Attention Mechanism, in advancing protein structure prediction and facilitating protein structure–function studies.