摘要
异方差线性判别分析(HLDA)因在语音识别中起到了巨大的特征去相关作用而被广泛利用。然而在训练数据不足或特征维数较高时,HLDA易出现不稳定性和小样本问题。根据特征的矩阵表示形式,提出了一种结构受限的HLDA。首先用二维线性判别分析(2DLDA)压缩矩阵形式的特征,然后作一维的HLDA。通过分析我们指出,二维的特征变换实际上是一种结构受限的一维特征变换。在RM库上的实验,受限HLDA对常规HLDA的词识别错误相对下降12.39%;在TIMIT库上的实验,受限HLDA对常规HLDA的音素识别错误相对下降4.43%。
Heteroscedastic linear discriminant analysis （HLDA） is applied widely in speech recognition due to its ability of feature de-correlation. To overcome its instability on high dimension features and the small sample issue on insufficient training samples, this paper proposes a structure-specific HLDA method to transform the feature matrix. The method adopts the two-dimensional linear discriminant analysis （2DLDA） to compress features in the matrix, and then, the one-dimensional HLDA is applied. It is revealed that two dimensional feature transformation is actually a structure-constrained one dimensional feature transformation. Experiments show that the proposed struc ture-specific HLDA achieves 12. 39% word error rate （WER） reduction on RM database and 4. 43% phone error rate （PER） reduction on TIMIT database compared with the traditional HLDA.
出处
《中文信息学报》
CSCD
北大核心
2008年第4期94-99,共6页
Journal of Chinese Information Processing
基金
国家863计划资助项目(2004AA114030)
关键词
计算机应用
中文信息处理
语音识别
特征变换
HLDA
结构受限
computer application
Chinese information processing
speech recognition
feature transformation
HLDA
structure-specific
