Open AccessImproved algorithm of DTW in speech recognition
Author(s) -
HU Zhi-Qiang,
Jiaqi Zhen,
Xin Wang,
Ziwei Liu,
Yong Liu
Publication year - 2019
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/563/5/052072
Subject(s) - dynamic time warping , computer science , path (computing) , distortion (music) , matlab , speech recognition , algorithm , pattern recognition (psychology) , search algorithm , limit (mathematics) , artificial intelligence , mathematics , amplifier , computer network , mathematical analysis , bandwidth (computing) , programming language , operating system
This paper improves the Dynamic Time Warping (DTW) algorithm. In order to change the problem of the traditional search range of DTW speech recognition algorithm is too large, an improved DTW algorithm is proposed to limit the search path. Firstly, the traditional algorithm is analyzed to find its speech recognition search path, distortion and recognition efficiency. Secondly, the improved algorithm is introduced, the search path is limited, and simulation is carried out in MATLAB. Compare the improvement results with the improved recognition results before comparing the distortion and recognition efficiency, and the improved efficiency is calculated. Experimental results show that the improved new speech recognition algorithm is superior to the traditional algorithm in overall performance.