Detecting cardiac diseases is crucial due to their widespread prevalence globally. Auscultation is the first step in diagnosing, and professionals depend primarily on their knowledge and experience in this area. Automating this process using artificial intelligence (AI) could significantly improve efficiency. AI has demonstrated the potential to enhance the accuracy of heart sound classification by analyzing relevant features and learning their relationships with different cardiac diseases. The 2016 PhysioNet/CinC Challenge database, which contains cardiac sound recordings, is one of the publicly accessible datasets used in this research. The developed model achieves a maximum accuracy of 91.04%. Furthermore, the study extends its contribution by employing Explainable Artificial Intelligence (XAI) to elucidate model predictions, enhancing interpretability. Notably, the novelty lies in the CNN model’s ability to classify heart sounds by integrating audio-specific features for improved classification performance.