Skip to content

Publications

Detail of publication

Citation

Matoušek, J and Tihelka, D. : SVM-Based Detection of Misannotated Words in Read Speech Corpora . Text, Speech and Dialogue, Proceedings of the 16th International Conference TSD 2013, Lecture Notes in Artificial Intelligence, vol. 8082, p. 457-464, Springer, Berlin-Heidelberg, Germany, 2013.

Additional information


Springerlink

Abstract

Automatic detection of misannotated words in single-speaker read-speech corpora is investigated in this paper. Support vector machine (SVM) classifier was proposed to detect the misannotated words. Its performance was evaluated with respect to various word-level feature sets. The SVM classifier was shown to perform very well with both high precision and recall scores and with F1 measure being almost 88%. This is a statistically significant improvement over a traditionally used outlier-based detection method.

Detail of publication

Title: SVM-Based Detection of Misannotated Words in Read Speech Corpora
Author: Matoušek, J ; Tihelka, D.
Language: English
Date of publication: 5 Sep 2013
Year: 2013
Type of publication: Papers in journals
Book title: Text, Speech and Dialogue, Proceedings of the 16th International Conference TSD 2013
Series: Lecture Notes in Artificial Intelligence
Číslo vydání: 8082
Page: 457 - 464
DOI: 10.1007/978-3-642-40585-3_58
ISSN: 0302-9743
Publisher: Springer
Address: Berlin-Heidelberg, Germany
Date: 1 Sep 2013 - 5 Sep 2013
/ 2014-01-26 22:36:11 /

Keywords

annotation error detection, classification, support vector machine, read speech corpora

BibTeX

@INCOLLECTION{MatousekJ_2013_SVM-BasedDetection,
 author = {Matou\v{s}ek, J and Tihelka, D.},
 title = {SVM-Based Detection of Misannotated Words in Read Speech Corpora},
 year = {2013},
 publisher = {Springer},
 address = {Berlin-Heidelberg, Germany},
 volume = {8082},
 pages = {457-464},
 booktitle = {Text, Speech and Dialogue, Proceedings of the 16th International Conference TSD 2013},
 series = {Lecture Notes in Artificial Intelligence},
 ISSN = {0302-9743},
 doi = {10.1007/978-3-642-40585-3_58},
 url = {http://www.kky.zcu.cz/en/publications/MatousekJ_2013_SVM-BasedDetection},
}