For two speeches (usually one is a native's utterance, and the other is a non-native learner's utterance), this method visualizes speech signal and its most relevant parameters and compares them. Moreover, auditory confirmation is also accelerated for feedback. Ongoing research in L2 acquisition has pointed out the learners' failure to perceive and produce difficult and new contrasts in the target language by ear alone. The use of visual information as an alternative feedback in the L2 learning is more effective[28]. On each trial, the target word is presented on the screen in orthographic form, and then the model speech is played over headphones or speakers. The learners produce the word by imitating model speech, and immediately after this production they can compare the results with the model parameters[28]. The goal is to help a supervisor or a leaner in identifying the most severe deviations in his (or her) pronunciation. Its main training aid is based on speech signal visualization and visual comparison. It is attractive, definitive, and the learner can easily comprehend the deviation, but it's too simple to understand how to correct the problem.
