Research on object assessment for mandarin proficiency test

Now the national mandarin proficiency test (NMPT) is in the way of manual work. In order to remove the human factors improve efficiency, NMPT based on computer-aided is expected urgently. In this paper, the novel Mandarin objective pronunciation assessment pronunciation of is proposed. The mandarin pronunciation is divided into consonant, vowel and tone. The standard of Mandarin pronunciation is divided into six levels. And the feature parameters are explored, and Bark-SD is selected as consonant assessment parameter. Hidden Markov Model is used to assess the pronunciation. The objective assessments correlate well with the subjective assessment because perception distortions of human auditory have been taken into consideration. Experimental results show the method performs better than the exiting methods.