This paper describes the characteristics of smart speaker and how and why this device is expected to play new role in language learning. Next we describe the characteristics of words called “onomatopoeia”, and the difficulties in learning them. We build a prototype of the application for smart speakers to learn Japanese onomatopoeia, implemented as a simple quiz application that plays the audio file as the question and requests the learner to answer what onomatopoeia to be used to express the sound. Then we discuss the findings from the prototype, and the future vision of the application of smart speakers to be combined with other learning assistant/management systems.
