Abstract: English education in Japan doesn’t provide enough time for students to be able to learn how to speak or communicate effectively. To solve this problem, we have been developing the Virtual Interview System (VIS) that virtualizes an interview test between students and teachers, increasing speaking practice times in language class. The VIS focuses on speaking and listening, but nonverbal communication, such as gestures, facial expression, and tone of voice are just as important. However, unlike spelling and punctuation it is very subjective. How can we assess the quality? In this report, we will highlight our preliminary research into how to quantify nonverbal communication: specifically with regards to facial expression. We used the Kinect 3D motion sensor camera to identify and quantify the quality or richness of facial expression, in our effort to help learners become better speakers and communicators of English.