文献类型：期刊文献

中文题名：DTW语音识别算法研究与分析

英文题名：The Research and Analysis of DTW speech recognition algorithm

作者：朱淑琴[1];赵瑛[1,2]

第一作者：朱淑琴

机构：[1]北京联合大学师范学院;[2]北京理工大学计算机学院

第一机构：北京联合大学师范学院

年份：2012

期号：5

起止页码：150-151

中文期刊名：微计算机信息

外文期刊名：Microcomputer Information

语种：中文

中文关键词：语音识别;动态时间规整算法;局部路径约束;区域限制;宽度限制

外文关键词：speech recognition; Dynamic Time Warping; local path constraint; area restrictions; width limit

摘要：研究动态时间规整(Dynamic Time Warping)语音识别算法问题,传统动态时间规整方法需要存储较大的矩阵,直接计算将会占据较大的空间,计算量也比较大,对系统硬件要求比较高。为了减小DTW算法的运算量,提高识别速度,对DTW语音识别算法进行优化改进。将局部路径约束和整体路径约束相结合,仅在一个规定的宽度内搜索动态规划路径,计算累积匹配距离。仿真实验结果表明该方法不仅可以降低运算负载,提高识别速度,而且能在一定程度上提高语音识别率。
Speech recognition based on dynamic time warping algorithm is discussed in the paper. The traditional DTW dynamic searching method needs large matrix storing, large space, a great of calculation and higher hardware demanding. In order to reduce the computation of the DTW algorithm and improve the recognition speed, DTW speech recognition algorithm is optimized and im- proved. Considering the local path constraint and the overall path constraint at the same time, speech recognition is carried out under the hmitation of search width. A large number of experiments show that the method can reduce the computational load, increase the recognition speed, but also to a certain extent, improve the speech recognition rate.