INTERSPEECH 2006 - ICSLP
This paper studies pronunciation scoring algorithm in CALL system aiming at teaching native Chinese learn standard Mandarin. Most of the pronunciation scoring algorithms focus on non-native environment, which may not be suitable for native speakers. We bring up a new algorithm based on traditional posterior log-likelihood algorithm by weighting the initial part of Mandarin syllables, where final-initialís duration ratio is introduced to control the weight. Experiments show that the proposed algorithm is much more effective than traditional posterior log-likelihood algorithm in the Mandarin learning system. The correlation with human score achieves an increase of 11%.
Bibliographic reference. Wei, Si / Liu, Qing-Sheng / Hu, Yu / Wang, Ren-Hua (2006): "Automatic Mandarin pronunciation scoring for native learners with dialect accent", In INTERSPEECH-2006, paper 1669-Tue3CaP.8.