Senone log-likelihood ratios based articulatory features in pronunciation erroneous tendency detecting
Leyuan Qu, Yanlu Xie, Jinsong Zhang · 2016
It is important to provide detailed and instructive feedback in computer assisted pronunciation training (CAPT) system. However the feedback is limited to the accuracy of the erroneous tendency detection. This paper proposed to apply senone log-likelihood ratio based articulatory features (AFs) to improve pronunciation erroneous tendency (PET) detection performance. Also the feedback information of articulation-placement and articulation-manner could be derived from the definition of PET. The framework of the method involved two main steps: (a) A bank of attribute extractors based on neural networks were trained to estimate the log-likelihood ratio (LLR) for each senone at a frame level; (b) AFs composed of those LLRs outputted from each attribute extractor were then used for detecting PETs. Results demonstrated that the system using AFs had better performance than the baseline system using MFCC. Moreover, substantial improvements were obtained by combining AFs with MFCC, achieving a lower false rejection rate of 5.0%, a lower false acceptance rate of 30.8% and a higher diagnostic accuracy of 89.8%.