Speech Recognition System Based Improved DTW Algorithm
Introduced the design and implementation of isolated word speech recognition system with application of the improved dynamic time warping (DTW) algorithm. Experimental results show that the use of the improved DTW algorithm effectively reduces the amount of data to be processed and the recognition time, improved the system speed. The algorithm showed more obvious advantage as the number of speech signal to be recognized increased.
DTW speech recognition endpoint detection training search path
Zhang Jing Zhang Min
Department of Computer Science and Technology Guangdong University of Foreign Studies Guangzhou, Chi Guangzhou Haige Communications Group Guangzhou Radio Group Guangzhou, China
国际会议
长春
英文
320-323
2010-08-24(万方平台首次上网日期,不代表论文的发表时间)