This paper proposes an improved context-aware language supported system for learning Japanese mimetic words and onomatopoeia (MIO) using wireless sensor network "MOTE". In our previous work, a learner wears sensors so that the system provides the appropriate MIO expressions at his/her place. In this current system called JAMIOLAS (Supporting Japanese Mimetic words and Onomatopoeia Learning with Sensors for Non-Japanese) 2.0, MOTE is used to get appropriate data from the environment, for example, temperature and light. The system recommends the right place for the right person to learn the right MIO expressions. This paper describes the implementation, the interface and usage scenario of JAMIOLAS 2.0.