Cardiovascular disease increasing deaths worldwide also it is affecting young age people at their early stage. Heartbeat analysis of a person can be normal or abnormal heart sounds which can be detected only through trained physician. To reduce the dependency on trained physicians for heart sound detection, proposed system focuses on automatic classification of PCG (Phonocardiogram) signals after removal of noise using Convolution Neural Network. Thus an automated machine/ artificial intelligence based preliminary analysis eliminates the need of trained physicians for detection of abnormalities in heart sounds. This paper proposes noise removal using spectral gating, Deep Learning based classification and energy based segmentation using to detect cardiac disease with efficient accuracy, specificity, and sensitivity. The proposed system becomes a support system for the physicians in automatic classification and diagnosis of the cardiovascular disease fast and efficient