Abstract- Phonocardiogram (PGC) signal is a graphical representation of the hearth sounds that can be used to diagnose a hearth disease. Diagnosing heart disease based on PGC signal is component including S1 and S2. Nevertheless, the interpretation of PGC signal is depend on the cardiologist's expertise. Therefore automated PGC signal classification is required in order to help the cardiologist diagnosing and monitoring heart disease. The classification of PGC signal is influenced by the segmentation and the feature extraction process. The segmentation process aims to detect the location of hearth sound components including S1 and S2 in PGC signal. However is difficult to find those component in a noisy PGC signal. The feature extraction process aims to extract relevant features that lie an segmented PGC signal. This process is required because the segmented PGC signal has high dimensionality and redundant information. This study proposes Shannon Energy Envelope for segmenting PCG signal and Deep Belief Network (DBN) for feature extraction method. The results show that the proposed method outperforms shallow models in existing datasets.