Poster Abstract: Mobile Context Logger: Recognizing User's Auditory Environments and Activities using Smartwatch
Akio Sashima, Mitsuru Kawamoto · 2023
Smartwatch, a wrist-worn personal device, can be a mobile platform for information assistant services that provide useful information in daily life, such as context-aware reminders for elderly people. To realize such an assistant service, we have developed a mobile context logger (MCL) that recognizes the user's auditory environments and activities based on auditory and motion information derived from the user's smartwatch. The MCL uses two embedded deep neural network modules: an audio recognition module and a motion recognition module. The modules were trained with open datasets and embedded on a smartwatch to recognize the user's auditory environments and activities. In this paper, we describe an overview of the MCL and discuss the issues towards the mobile context-aware assistant service based on our development.