摘要
提出了一种基于模糊C-means聚类法的矢量量化,并将其用于语音特征的矢量量化。从语音信号中提取12阶MFCC倒谱系数作为待分群样本的矢量数据,有效地降低数据量及计算量,并可以避免杂信的不良影响。且实验得到的码本分布合理,没有空类,表明了该量化方法对语音识别很有效。
This paper presented a method based on fuzzy c - means clustering, which is applied to vector quantization of speech feature. This algorithm extracts 12th order MFCC form speech signals and makes them as the vector data,which will be classified. This method can effectively reduce＇ the volume of data and computing, and avoid the adverse effects of impurities. The experiment result confirmed that distribution of the code vector that is obtained by fuzzy C - means clustering is reasonable and there are not empty classes, and this kind of code vector demonstrate the efficiency of this quantization method for speech recognition.
出处
《微计算机应用》
2007年第11期1164-1168,共5页
Microcomputer Applications
