In the past few decades, analysis of heart sound signals (i.e., the phonocardiogram or PCG), especially for automated heart sound segmentation and classification, has been widely studied and has been reported to have the potential value to detect pathology accurately in clinical applications. However, the current electronic auscultation device can only record and store heart sound data and without any Al-based analysis and clinic-condition functions.So we develop the smart heart sound device. This device is based on the NVIDIA Jetson TX2 and we build AI models with combination of CNN and RNN. For data, we use the data from Physionet/CinC Challenge 2016 to train the model. Eventually the smart heart sound device can record and appear people’s heart sound signal, automatically analyze the signal to see whether is normal/abnormal and feed back to users. In addition, the device can be accessed to IoT system, so the users’ data is able to be sent to family doctors immediately and doctors will find out the specific disease.