自动发音错误检测中基于最大化F1值准则的区分性特征补偿训练算法  

Maximum F1-Score Criterion Based Discriminative Feature Compensation Training Algorithm for Automatic Mispronunciation Detection

作　　者：黄浩[1] 徐海华[2] 王羡慧[1] 吾守尔.斯拉木 

机构地区：[1]新疆大学信息科学与工程学院,新疆乌鲁木齐830046 [2]南洋理工大学Temasek实验室,新加坡639798

出　　处：《电子学报》2015年第7期1294-1299,共6页Acta Electronica Sinica

基　　金：国家自然科学基金(No.61365005;No.60965002)

摘　　要：为提高自动发音错误检测性能,提出一种区分性特征补偿训练算法.该方法将高斯后验概率矢量经过线性变换后作为偏移量补偿至传统的谱特征.将经过正确度标注的语音数据库上的发音错误检测F1值的最大化作为变换参数的训练准则.推导了目标函数对变换参数的偏导数公式,并利用无约束参数优化例程L-BFGS更新变换参数.发音错误检测实验表明该方法能够有效增大训练和测试集的F1值.并且训练和测试集的精确度、召回率也都有明显提高.在特征优化的基础上进行模型参数训练,检错性能较单独的区分性特征训练、单独的区分性模型训练都有进一步改进.To improve the performance of automatic mispronunciation detection,a discriminative feature compensation training algorithm is proposed. The method is to train a matrix projecting from posteriors of Gaussians to a normal size feature space,and then to add the projected features to traditional spectral features. The matrix is trained according to maximum F1-score criterion,which aims at maximizing the empirical mispronunciation detection F1-score on the annotated speech database. Mispronunciation detection experiments have shown the method is effective in increasing F1-score,precision and recall on both the training data and evaluation data. It is also shown model parameter discriminative training on newfeatures obtained further improvements over both model training and feature training.

关 键 词：自动发音错误检测 F1值 区分性训练 特征 计算机辅助语言学习 

分 类 号：TN912.34[电子电信—通信与信息系统]