Paper Info
Title
Lexical Stress-Based Morphological Decomposition and Its Application for Ukrainian Speech Recognition.
Abstract
This paper presents an approach to word morphological decomposition based on lexical stress modeling. Word segmentation quality is estimated by a hidden variable that assigns the lexical stress. The formulated segmentation criterion is based on a training set of words with manually pointed stresses and a large text corpus. The described search algorithm finds one or more segmentations with the best likelihood. Given arguments confirm the necessity to distinguish stressed and unstressed vowels in the phoneme alphabet for Ukrainian speech recognition systems. The developed tool allows to assign primary lexical stress in unknown words. Experimental research is described and results are discussed.
Year
DOI
Venue
2013
10.1007/978-3-642-40585-3_42
Lecture Notes in Computer Science
Keywords
Field
DocType
lexical stress,morphological decomposition,speech recogntition,Ukrainian
Training set,Search algorithm,Computer science,Segmentation,Text corpus,Speech recognition,Text segmentation,Ukrainian,Artificial intelligence,Natural language processing,Hidden variable theory,Alphabet
Conference
Volume
ISSN
Citations 
8082
0302-9743
0
PageRank 
References 
Authors
0.34
5
2
Name
Order
Citations
PageRank
Mykola M. Sazhok101.01
Valentyna Robeiko201.01