Pronunciation Quality Scoring for Single Syllable Word in PSC

This paper discusses pronunciation quality scoring for single syllable word in Putonghua Shuiping Ceshi (PSC) that is a nationwide spoken test to evaluate the standard level and the practical ability that an individual uses the mandarin in china. This study mainly includes some algorithms about the syllable separation, acoustic units selection, posterior probability scoring, threshold values setting and neural network combination. Experiment shows that the proposed approach achieves a high correlation of 0.731, a value very close to 0.786 between human experts. It is also observed that the combination method of neural network gets the best evaluation result. This method is eligible for the automatic scoring in PSC.

[1]  Yoon Kim,et al.  Automatic pronunciation scoring for language instruction , 1997, 1997 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[2]  Yu Shi,et al.  Segmental tonal modeling for phone set design in Mandarin LVCSR , 2004, 2004 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[3]  Mitch Weintraub,et al.  Automatic scoring of pronunciation quality , 2000, Speech Commun..

[4]  Frank K. Soong,et al.  Generalized Segment Posterior Probability for Automatic Mandarin Pronunciation Evaluation , 2007, 2007 IEEE International Conference on Acoustics, Speech and Signal Processing - ICASSP '07.

[5]  Silke M. Witt,et al.  Use of speech recognition in computer-assisted language learning , 2000 .