Paper Info
Title
Improving the Accuracy of Large Vocabulary Continuous Speech Recognizer Using Dependency Parse Tree and Chomsky Hierarchy in Lattice Rescoring
Abstract
This research work describes our approaches in using dependency parse tree information to derive useful hidden word statistics to improve the baseline system of Malay large vocabulary automatic speech recognition system. The traditional approaches to train language model are mainly based on Chomsky hierarchy type 3 that approximates natural language as regular language. This approach ignores the characteristics of natural language. Our work attempted to overcome these limitations by extending the approach to consider Chomsky hierarchy type 1 and type 2. We extracted the dependency tree based lexical information and incorporate the information into the language model. The second pass lattice rescoring was performed to produce better hypotheses for Malay large vocabulary continuous speech recognition system. The absolute WER reduction was 2.2% and 3.8% for MASS and MASS-NEWS Corpus, respectively.
Year
DOI
Venue
2013
10.1109/IALP.2013.53
IALP
Keywords
Field
DocType
lvcsr,large vocabulary continuous speech,chomsky hierarchy type,lattice rescoring,speech recognition,natural language,mass-news corpus,trees (mathematics),dependency tree,recognition system,second pass lattice rescoring,language model,large vocabulary continuous speech recognizer,mass corpus,dependency parse tree information,dependency tree based lexical information,malay recognizer,lexical information,hidden word statistics,large vocabulary automatic speech,chomsky hierarchy,malay large vocabulary automatic speech recognition system,regular language,linguistic information,baseline system,natural language processing,absolute wer reduction,dependency parse tree
Rule-based machine translation,Parse tree,Computer science,Malay,Chomsky hierarchy,Speech recognition,Natural language,Natural language processing,Artificial intelligence,Regular language,Vocabulary,Language model
Conference
Citations 
PageRank 
References 
0
0.34
0
Authors
3
Name
Order
Citations
PageRank
Kai Sze Hong111.42
Tien-Ping Tan2257.46
Enya Kong Tang354.50