Exploring the automatic mispronunciation detection of confusable phones for mandarin

Mispronunciation detection is one of the vital tasks of the CALL (Computer Assisted Language Learning) systems. Many methods have been introduced to accomplish this task. However, few of them have addressed the detection task on confusable phones. In this paper, phone-level classifiers are utilized to improve the detection performance on the confusable phones. Features of the classifiers are posterior probability vectors calculated from their corresponding acoustic models. Moreover, confusion matrix is also extracted and incorporated to calculate derivatives of the posterior probability vectors. Experiments on our Mandarin database validate the effectiveness of our proposed method, compared with the commonly used posterior probability and phone dependent thresholds methods.