DocumentCode :
2261934
Title :
Large vocabulary word recognition based on a graph-structured dictionary
Author :
Minamino, Katsuki
Author_Institution :
Sony Corp., Tokyo, Japan
Volume :
4
fYear :
1996
fDate :
3-6 Oct 1996
Firstpage :
2123
Abstract :
In this paper, a structural search using a word-node graph is proposed to speed up the isolated word recognition based on hidden Markov models (HMMs). We define a distance measure for comparing pairs of words, and construct a graph containing the word distribution structure. Based on this graph, the number of words to be examined are restricted. Experiments show that the search complexity is considerably reduced with little degradation of the recognition accuracy
Keywords :
computational complexity; graph theory; hidden Markov models; search problems; speech recognition; vocabulary; distance measure; graph-structured dictionary; hidden Markov models; isolated word recognition; large vocabulary word recognition; recognition accuracy degradation; search complexity; structural search; word distribution structure; word-node graph; Degradation; Dictionaries; Hidden Markov models; Laboratories; Probability density function; Real time systems; Speech recognition; Vector quantization; Viterbi algorithm; Vocabulary;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Spoken Language, 1996. ICSLP 96. Proceedings., Fourth International Conference on
Conference_Location :
Philadelphia, PA
Print_ISBN :
0-7803-3555-4
Type :
conf
DOI :
10.1109/ICSLP.1996.607222
Filename :
607222
Link To Document :
بازگشت