Researchers have made a lot of effort for the lightweight and high flexibility of wearable electronic devices, which also requires the associated energy harvesting equipment to have ultra-thin thickness and high stretchability. Therefore, a piezoelectric-triboelectric hybrid self-powered sensor (PTHS) has been proposed which can be used as the second layer of the human body. This elastic PTHS can even work on a person's fingers without disturbing the body's movements. The open circuit voltage and short circuit current of devices with a projected area of 30 mm × 25 mm can reach 1.2 V and 30 nA, respectively. Two piezoelectrically-triboelectrically sensors with machine learning optimized identification strategies were experimentally proven as the potential applications of the PTHS. The PTHS's ultra-thin thickness, high stretchability and superior geometry control features are promising in electronic skin, artificial muscle and soft robotics. The novelty of this work is that a smart mask integrated with PTHS can generate a signal of the hybrid sensor for the biomechanical motion classifier. After suitable training, an overall accuracy of 87.9% using long short-term memory can be achieved.