Click to open the HelpDesk interface
AECE - Front page banner

Menu:


FACTS & FIGURES

JCR Impact Factor: 0.595
JCR 5-Year IF: 0.661
Issues per year: 4
Current issue: Aug 2017
Next issue: Nov 2017
Avg review time: 77 days


PUBLISHER

Stefan cel Mare
University of Suceava
Faculty of Electrical Engineering and
Computer Science
13, Universitatii Street
Suceava - 720229
ROMANIA

Print ISSN: 1582-7445
Online ISSN: 1844-7600
WorldCat: 643243560
doi: 10.4316/AECE


TRAFFIC STATS

1,718,652 unique visits
505,892 downloads
Since November 1, 2009



Robots online now
Baiduspider


SJR SCImago RANK

SCImago Journal & Country Rank


SEARCH ENGINES

aece.ro - Google Pagerank




TEXT LINKS

Anycast DNS Hosting
MOST RECENT ISSUES

 Volume 17 (2017)
 
     »   Issue 3 / 2017
 
     »   Issue 2 / 2017
 
     »   Issue 1 / 2017
 
 
 Volume 16 (2016)
 
     »   Issue 4 / 2016
 
     »   Issue 3 / 2016
 
     »   Issue 2 / 2016
 
     »   Issue 1 / 2016
 
 
 Volume 15 (2015)
 
     »   Issue 4 / 2015
 
     »   Issue 3 / 2015
 
     »   Issue 2 / 2015
 
     »   Issue 1 / 2015
 
 
 Volume 14 (2014)
 
     »   Issue 4 / 2014
 
     »   Issue 3 / 2014
 
     »   Issue 2 / 2014
 
     »   Issue 1 / 2014
 
 
  View all issues  


FEATURED ARTICLE

Wind Speed Prediction with Wavelet Time Series Based on Lorenz Disturbance, ZHANG, Y., WANG, P., CHENG, P., LEI, S.
Issue 3/2017

AbstractPlus






LATEST NEWS

2017-Jun-14
Thomson Reuters published the Journal Citations Report for 2016. The JCR Impact Factor of Advances in Electrical and Computer Engineering is 0.595, and the JCR 5-Year Impact Factor is 0.661.

2017-Apr-04
We have the confirmation Advances in Electrical and Computer Engineering will be included in the EBSCO database.

2017-Feb-16
With new technologies, such as mobile communications, internet of things, and wide applications of social media, organizations generate a huge volume of data, much faster than several years ago. Big data, characterized by high volume, diversity and velocity, increasingly drives decision making and is changing the landscape of business intelligence, from governments to private organizations, from communities to individuals. Big data analytics that discover insights from evidences has a high demand for computing efficiency, knowledge discovery, problem solving, and event prediction. We dedicate a special section of Issue 4/2017 to Big Data. Prospective authors are asked to make the submissions for this section no later than the 31st of May 2017, placing "BigData - " before the paper title in OpenConf.

2017-Jan-30
We have the confirmation Advances in Electrical and Computer Engineering will be included in the Gale database.

2016-Dec-17
IoT is a new emerging technology domain which will be used to connect all objects through the Internet for remote sensing and control. IoT uses a combination of WSN (Wireless Sensor Network), M2M (Machine to Machine), robotics, wireless networking, Internet technologies, and Smart Devices. We dedicate a special section of Issue 2/2017 to IoT. Prospective authors are asked to make the submissions for this section no later than the 31st of March 2017, placing "IoT - " before the paper title in OpenConf.

Read More »


    
 

  1/2011 - 13

Domain Independent Vocabulary Generation and Its Use in Category-based Small Footprint Language Model

KIM, K.-H. See more information about KIM, K.-H. on SCOPUS See more information about KIM, K.-H. on IEEExplore See more information about KIM, K.-H. on Web of Science, KIM, J.-H. See more information about KIM, J.-H. on SCOPUS See more information about KIM, J.-H. on SCOPUS See more information about KIM, J.-H. on Web of Science
 
Click to see author's profile on See more information about the author on SCOPUS SCOPUS, See more information about the author on IEEE Xplore IEEE Xplore, See more information about the author on Web of Science Web of Science

Download PDF pdficon (639 KB) | Citation | Downloads: 1,089 | Views: 2,507

Author keywords
natural language processing, speech recognition

References keywords
language(16), speech(12), spoken(5), recognition(5), processing(5), modeling(5), model(5), vocabulary(4), statistical(4), gram(4)
Blue keywords are present in both the references section and the paper title.

About this article
Date of Publication: 2011-02-27
Volume 11, Issue 1, Year 2011, On page(s): 77 - 84
ISSN: 1582-7445, e-ISSN: 1844-7600
Digital Object Identifier: 10.4316/AECE.2011.01013
Web of Science Accession Number: 000288761800013
SCOPUS ID: 79955973325

Abstract
Quick view
Full text preview
The work in this paper pertains to domain independent vocabulary generation and its use in category-based small footprint Language Model (LM). Two major constraints of the conventional LMs in the embedded environment are memory capacity limitation and data sparsity for the domain-specific application. This data sparsity adversely affects vocabulary coverage and LM performance. To overcome these constraints, we define a set of domain independent categories using a Part-Of-Speech (POS) tagged corpus. Also, we generate a domain independent vocabulary based on this set using the corpus and knowledge base. Then, we propose a mathematical framework for a category-based LM using this set. In this LM, one word can be assigned assign multiple categories. In order to reduce its memory requirements, we propose a tree-based data structure. In addition, we determine the history length of a category n-gram, and the independent assumption applying to a category history generation. The proposed vocabulary generation method illustrates at least 13.68% relative improvement in coverage for a SMS text corpus, where data are sparse due to the difficulties in data collection. The proposed category-based LM requires only 215KB which is 55% and 13% compared to the conventional category-based LM and the word-based LM, respectively. It successively improves the performance, achieving 54.9% and 60.6% perplexity reduction compared to the conventional category-based LM and the word-based LM in terms of normalized perplexity.


References | Cited By  «-- Click to see who has cited this paper

[1] S. Young, "A Reivew of Large Vocabulary Continuous Speech Recognition," IEEE Signal Processing Magazine, vol. 13, no. 5, pp. 45-57, 1990.
[CrossRef] [SCOPUS Times Cited 218]


[2] S. Young, G. Evermann, D. Kershaw, G. Moore, J. Odell, D. Ollason, D. Povey, V. Valtchev, and P. Woodland, The HTK Book (for HTK version 3.2), Cambridge University Engineering Department, 2002.

[3] K. Lee, H. Hon, and R. Reddy, "An Overview of the SPHINX Speech Recognition," IEEE Transactions on Acoustics, Speech, and Signal Processing, vol. 38, no. 1, pp. 35-45, 1990.
[CrossRef] [Web of Science Times Cited 103] [SCOPUS Times Cited 163]


[4] J. Novak, P. Dixon, and S. Furui, "An Empirical Comparison of the T3, Juicer, HDecode and Sphinx3 Decoders," Proc. Interspeech, pp.1890-1893, 2010.

[5] M. Adda-Decker and L. Lamel, "The Use of Lexica in Automatic Speech Recognition," Lexicon Development for Speech and Language Processing, F. van Eynde, D. Gibbon (Eds.), Kluwer Academic, pp. 235-266, 2000.

[6] R. Rosenfeld, "Optimizing Lexical and N-gram Coverage via Judicious Use of Linguistic Data," Proc. Eurospeech, pp. 1763-1766, 1995.

[7] S. Adolphs and N. Shemitt, "Lexical Coverage of Spoken Discourse," Applied Linguistics, vol. 24, no. 4, pp. 425-438, 2003.
[CrossRef] [Web of Science Times Cited 55] [SCOPUS Times Cited 59]


[8] P. Nation and R. Waring, "Vocabulary Size, Text Coverage and Word Lists," Vocabulary: Description, Acquisition and Pedagogy, N. Schmitt, M. McCarthy (Eds.), Cambridge University Press, pp. 6-19, 1997.

[9] S. Katz, "Estimation of Probabilities from Sparse Data for The Language Model Component of a Speech Recognizer," IEEE Transaction on Acoustic, Speech and Signal Processing., vol. 35, no. 3, pp. 400-401, 1987.
[CrossRef] [Web of Science Times Cited 432] [SCOPUS Times Cited 693]


[10] F. Bechet, Y. Esteve, and R. Mori, "Tree-based Language Model Dedicated to Natural Spoken Dialog System," Proc. International Symposium on Computer Architecture, pp. 207-210, 2001.

[11] C. Troncoso and T. Kwawahara, "Trigger-based Language Model Adaptation for Automatic Meeting Transcription," Proc. Interspeech., pp. 1297-1300, 2005.

[12] I. Zitouni, K. Samili, and J. Haton, "Statistical Language Modeling Based on Variable-length Sequence," Computer Speech and Language, vol. 17, no. 1, pp. 27-41, 2003.
[CrossRef] [Web of Science Times Cited 8] [SCOPUS Times Cited 11]


[13] I. Zitouni, "Backoff Hierarchical Class N-gram Language Models: Effectiveness to Model Unseen Events in Speech Recognition," Computer Speech and Language., vol. 21, no. 1, pp.88-104, 2007.
[CrossRef] [Web of Science Times Cited 12] [SCOPUS Times Cited 23]


[14] H. Yamamoto, S. Isogai, and Y. Sagisaka, "Multi-class Composite N-gram Language Model," Speech Communication., vol. 41, no. 2, pp 369-379, 2003.
[CrossRef] [Web of Science Times Cited 26] [SCOPUS Times Cited 35]


[15] P. Brown, V. Pietra, P. deSouza, J. Lai, and L. Mercer, "Class-based n-gram Models of Natural Language," Computational Linguistics., vol. 18, no. 4, pp. 467-479, 1990.

[16] R. Kneser and H. Ney, "Improved Clustering Techniques for Class-based Statistical Language Modeling," Proc. Eurospeech., pp.973-976, 1993.

[17] J. Stig, "Word Frequency and Text Type: Some Observations based on the LOB Corpus of British English texts," Computers and the Humanities, vol. 19, no. 1, pp. 23-36, 1985.
[CrossRef] [Web of Science Times Cited 8] [SCOPUS Times Cited 11]


[18] C. Fellbaum, WordNet: An Electronic Lexical Database, MIT Press, 1998.

[19] G. Leech, P. Rayson, and A. Wilson, Word Frequencies in Written and Spoken English: Based on the British National Corpus, Pearson ESL, 2001.

[20] R. Ordelman, A. Hessen, and F. Jong, "Lexicon Optimization for Dutch Speech Recognition in Spoken Document Retrieval," Proc. Eurospeech, pp. 1085-1088, 2001.

[21] J. Burger, J. Henderson, and W. Morgan, "Statistical Named Entity Recognizer Adaptation," Proc. Natural Language Learning, pp. 1-4, 2002.

[22] T. Hasegawa and S. Sekine, "Discovering Relations Among Named Entities from Large Corpora," Proc. Association for Computational Linguistics, pp. 415-442, 2004

[23] H. Blockeel, L. Raedt, and J. Ramon, "Top-down Induction of Clustering Trees," Proc. International Conference on Machine Learning, pp.55-63, 1998.

[24] P. Banerjee and H. Han, "Language Modeling Approaches to Information Retrieval," Journal of Computing Science and Engineering, vol. 3, no. 3, pp. 143-164, 2009.

[25] N. Schmitt and M. McCarthy, Vocabulary: Description, Acquisition and Pedagogy, Cambridge University Press, pp. 6-19, 1997.

[26] P. Clarkson and R. Rosenfeld, "Statistical Language Modeling Using the CMU-Cambridge Toolkit," Proc. Eurospeech., pp. 2707-2710, 1997.

[27] A. Stolcke, "SRILM-an Extensible Language Modeling Toolkit," Proc. International Conference on Spoken Language Processing, pp. 901-904, 2002.

References Weight

Web of Science® Citations for all references: 644 TCR
SCOPUS® Citations for all references: 1,213 TCR

Web of Science® Average Citations per reference: 24 ACR
SCOPUS® Average Citations per reference: 45 ACR

TCR = Total Citations for References / ACR = Average Citations per Reference

We introduced in 2010 - for the first time in scientific publishing, the term "References Weight", as a quantitative indication of the quality ... Read more

Citations for references updated on 2017-09-22 21:03 in 58 seconds.




Note1: Web of Science® is a registered trademark of Thomson Reuters.
Note2: SCOPUS® is a registered trademark of Elsevier B.V.
Disclaimer: All queries to the respective databases were made by using the DOI record of every reference (where available). Due to technical problems beyond our control, the information is not always accurate. Please use the CrossRef link to visit the respective publisher site.

Copyright ©2001-2017
Faculty of Electrical Engineering and Computer Science
Stefan cel Mare University of Suceava, Romania


All rights reserved: Advances in Electrical and Computer Engineering is a registered trademark of the Stefan cel Mare University of Suceava. No part of this publication may be reproduced, stored in a retrieval system, photocopied, recorded or archived, without the written permission from the Editor. When authors submit their papers for publication, they agree that the copyright for their article be transferred to the Faculty of Electrical Engineering and Computer Science, Stefan cel Mare University of Suceava, Romania, if and only if the articles are accepted for publication. The copyright covers the exclusive rights to reproduce and distribute the article, including reprints and translations.

Permission for other use: The copyright owner's consent does not extend to copying for general distribution, for promotion, for creating new works, or for resale. Specific written permission must be obtained from the Editor for such copying. Direct linking to files hosted on this website is strictly prohibited.

Disclaimer: Whilst every effort is made by the publishers and editorial board to see that no inaccurate or misleading data, opinions or statements appear in this journal, they wish to make it clear that all information and opinions formulated in the articles, as well as linguistic accuracy, are the sole responsibility of the author.




Website loading speed and performance optimization powered by: