Conference paper

 
Kombrink, S., Hannemann, M., Burget, L.: Out-of-vocabulary word detection and beyond, In: ECML PKDD 2010 Proceedings and Journal Content, Barcelona, ES, 2010, p. 1-8
Publication language:english
Original title:Out-of-vocabulary word detection and beyond
Title (cs):Rozpoznávání slova mimo slovník a jeho hranice
Pages:1-8
Proceedings:ECML PKDD 2010 Proceedings and Journal Content
Conference:European Conference on Machine Learning and Priciples and Practice of Knowledge Discovery in Databases
Place:Barcelona, ES
Year:2010
URL:http://www.fit.vutbr.cz/research/groups/speech/publi/2010/kombrink_DIRAC_WS-1_2010.pdf [PDF]
Keywords
out-of-vocabulary word, detection, neural network, speech recognition
Annotation
In this work, we investigate into two approaches for OOV word detection. We compare both systems in a fusion experiment, and describe how to actually make use of the detected incongruence.
Abstract
In this work, we summarize our experiences in detection of unexpected words in automatic speech recognition (ASR). Two ap- proaches based upon a paradigm of incongruence detection between generic and specific recognition systems are introduced. By arguing, that detec- tion of incongruence is a necessity, but does not suffice when having in mind possible follow-up actions, we motivate the preference of one ap- proach over the other. Nevertheless, we show, that a fusion outperforms both single systems. Finally, we propose possible actions after the de- tection of unexpected words, and conclude with general remarks about what we found to be important when dealing with unexpected words.
BibTeX:
@INPROCEEDINGS{
   author = {Stefan Kombrink and Mirko Hannemann and Lukáš Burget},
   title = {Out-of-vocabulary word detection and beyond},
   pages = {1--8},
   booktitle = {ECML PKDD 2010 Proceedings and Journal Content},
   year = {2010},
   location = {Barcelona, ES},
   language = {english},
   url = {http://www.fit.vutbr.cz/research/view_pub.php?id=9373}
}