Use of an intermediate face between a learner and a teacher in second language learning with shadowing

Shadowing is a language-learning method whereby a learner attempts to repeat, i.e., shadow, what he/she hears immediately. We propose displaying a computer-generated intermediate face between a learner and a teacher as an appropriate intermediate scaffold for shadowing. The intermediate face allows the learner to follow a teacher's face and mouth movements more effectively. We describe a prototype system that generates an intermediate face from real-time camera input and captured video. We also discuss a user study of the prototype system with crowd-sourced participants. The results of the user study suggest that the prototype system provided better pronunciation cues than video-only shadowing techniques.