基于DIVA模型的英语辅音发音错误自动校正方法

Automatic correction method of English consonant pronunciation errors based on DIVA model

针对目前方法进行英语口语发音校正时,存在漏校正数量较高的问题,提出了基于DIVA模型的英语辅音发音错误自动校正方法。该方法首先依据DIVA模型(Directions into Velocities of Articulators Model)生成学习者语音辅音发音数据;结合HMM模型(Hidden Markov Model)建立学习者口语发音模型对数据进行训练提取数据特征;最后通过混淆的矩阵因素聚类算法计算学习者发音特征与正确发音数据之间的相似度,获取校正系数值,完成口语发音错误的校正。实验结果表明,运用该方法校正口语发音时的校正数量较高、校正成功率较好、漏校正数量较低。

To solve the problem of the high number of missed corrections when the current methods are used to correct the pronunciation of spoken English,an automatic correction method of English consonant pronunciation errors based on the DIVA model is proposed.This method first generates the pronunciation data of the learner’s voice consonants according to the DIVA model.Then,the HMM model is combined to build a learner’s spoken language pronunciation model to train the data to extract data features.Finally,the confusion matrix factor clustering algorithm is used to calculate the similarity between the learner’s pronunciation characteristics and the correct pronunciation data,thus,the correction coefficient value is obtained to complete the correction of the spoken pronunciation error.The experiment results show that when using this method to correct spoken pronunciation,the number of corrections is high,the correction success rate is good,and the number of missed corrections is low.