An Emergency Application for Smartphones Based on Rhythm Pattern Recognition

Smartphones are equipped with a variety of functions and are used by many people. Some of these functions can be effectively used in an emergency situation, such as a sound recording function for collecting evidence and a calling function for summoning help. However, touch-based operations characteristic of smartphones, which have no physical buttons, can sometimes be difficult. Operations using speech recognition are also possible, but sending out an SOS without another person noticing is difficult with this method. Thus, regardless of how convenient a smartphone function may be, using it in an emergency may be difficult. In this study, we propose an emergency application for smartphones based on rhythm pattern recognition. This approach enables operations to be performed without having to look at the screen, which means smartphone operations that can be used when the user does not want another person to notice or wants to keep another person's movements in view. This application is equipped with three functions: a remote SOS function for sending out an SOS to a person located elsewhere, an evidence recording function that records evidence by audio or video means, and a local SOS function that summons help from nearby people while intimidating the person posing a threat. In an experiment performed to evaluate this application, 70% of the subjects were able to input two rhythm patterns while carrying on a conversation without the other person noticing.