Modeling characteristics of agglutinative languages with multi-class language model for ASR system

I. Dawa; Y. Sagisaka; S. Nakamura

doi:10.1109/ICSDA.2009.5278368

Modeling characteristics of agglutinative languages with multi-class language model for ASR system

I. Dawa, Y. Sagisaka, S. Nakamura

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

1 Citation (Scopus)

Abstract

In this paper, we discuss a new language model that considers the characteristics of the agglutinative languages. We used Mongolian (a Cyrillic language system used in Mongolia) as an example from which to build the language model. We developed a Multi-class N-gram language model based on similar word clustering that focuses on the variable suffixes of a word in Mongolian. By applying our proposed language model, the resulting recognition system can improve performance by 6.85% compared with a conventional word N-gram when applying the ATRASR engine. We also confirmed that our new model will be convenient for rapid development of an ASR system for resource-deficient languages, especially for agglutinative languages such as Mongolian.

Original language	English
Title of host publication	2009 Oriental COCOSDA International Conference on Speech Database and Assessments, ICSDA 2009
Pages	104-109
Number of pages	6
DOIs	https://doi.org/10.1109/ICSDA.2009.5278368
Publication status	Published - 2009 Dec 10
Externally published	Yes
Event	2009 Oriental COCOSDA International Conference on Speech Database and Assessments, ICSDA 2009 - Urumqi, China Duration: 2009 Aug 10 → 2009 Aug 12

Publication series

Name	2009 Oriental COCOSDA International Conference on Speech Database and Assessments, ICSDA 2009

Conference

Conference	2009 Oriental COCOSDA International Conference on Speech Database and Assessments, ICSDA 2009
Country/Territory	China
City	Urumqi
Period	09/8/10 → 09/8/12

ASJC Scopus subject areas

Language and Linguistics
Software
Linguistics and Language

Access to Document

10.1109/ICSDA.2009.5278368

Cite this

Dawa, I., Sagisaka, Y., & Nakamura, S. (2009). Modeling characteristics of agglutinative languages with multi-class language model for ASR system. In 2009 Oriental COCOSDA International Conference on Speech Database and Assessments, ICSDA 2009 (pp. 104-109). Article 5278368 (2009 Oriental COCOSDA International Conference on Speech Database and Assessments, ICSDA 2009). https://doi.org/10.1109/ICSDA.2009.5278368

Modeling characteristics of agglutinative languages with multi-class language model for ASR system. / Dawa, I.; Sagisaka, Y.; Nakamura, S.
2009 Oriental COCOSDA International Conference on Speech Database and Assessments, ICSDA 2009. 2009. p. 104-109 5278368 (2009 Oriental COCOSDA International Conference on Speech Database and Assessments, ICSDA 2009).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Dawa, I, Sagisaka, Y & Nakamura, S 2009, Modeling characteristics of agglutinative languages with multi-class language model for ASR system. in 2009 Oriental COCOSDA International Conference on Speech Database and Assessments, ICSDA 2009., 5278368, 2009 Oriental COCOSDA International Conference on Speech Database and Assessments, ICSDA 2009, pp. 104-109, 2009 Oriental COCOSDA International Conference on Speech Database and Assessments, ICSDA 2009, Urumqi, China, 09/8/10. https://doi.org/10.1109/ICSDA.2009.5278368

Dawa I, Sagisaka Y, Nakamura S. Modeling characteristics of agglutinative languages with multi-class language model for ASR system. In 2009 Oriental COCOSDA International Conference on Speech Database and Assessments, ICSDA 2009. 2009. p. 104-109. 5278368. (2009 Oriental COCOSDA International Conference on Speech Database and Assessments, ICSDA 2009). doi: 10.1109/ICSDA.2009.5278368

@inproceedings{5a9d568a39a3423daf85ad43b585a8a3,

title = "Modeling characteristics of agglutinative languages with multi-class language model for ASR system",

abstract = "In this paper, we discuss a new language model that considers the characteristics of the agglutinative languages. We used Mongolian (a Cyrillic language system used in Mongolia) as an example from which to build the language model. We developed a Multi-class N-gram language model based on similar word clustering that focuses on the variable suffixes of a word in Mongolian. By applying our proposed language model, the resulting recognition system can improve performance by 6.85% compared with a conventional word N-gram when applying the ATRASR engine. We also confirmed that our new model will be convenient for rapid development of an ASR system for resource-deficient languages, especially for agglutinative languages such as Mongolian.",

author = "I. Dawa and Y. Sagisaka and S. Nakamura",

year = "2009",

month = dec,

day = "10",

doi = "10.1109/ICSDA.2009.5278368",

language = "English",

isbn = "9781424444007",

series = "2009 Oriental COCOSDA International Conference on Speech Database and Assessments, ICSDA 2009",

pages = "104--109",

booktitle = "2009 Oriental COCOSDA International Conference on Speech Database and Assessments, ICSDA 2009",

note = "2009 Oriental COCOSDA International Conference on Speech Database and Assessments, ICSDA 2009 ; Conference date: 10-08-2009 Through 12-08-2009",

}

TY - GEN

T1 - Modeling characteristics of agglutinative languages with multi-class language model for ASR system

AU - Dawa, I.

AU - Sagisaka, Y.

AU - Nakamura, S.

PY - 2009/12/10

Y1 - 2009/12/10

N2 - In this paper, we discuss a new language model that considers the characteristics of the agglutinative languages. We used Mongolian (a Cyrillic language system used in Mongolia) as an example from which to build the language model. We developed a Multi-class N-gram language model based on similar word clustering that focuses on the variable suffixes of a word in Mongolian. By applying our proposed language model, the resulting recognition system can improve performance by 6.85% compared with a conventional word N-gram when applying the ATRASR engine. We also confirmed that our new model will be convenient for rapid development of an ASR system for resource-deficient languages, especially for agglutinative languages such as Mongolian.

AB - In this paper, we discuss a new language model that considers the characteristics of the agglutinative languages. We used Mongolian (a Cyrillic language system used in Mongolia) as an example from which to build the language model. We developed a Multi-class N-gram language model based on similar word clustering that focuses on the variable suffixes of a word in Mongolian. By applying our proposed language model, the resulting recognition system can improve performance by 6.85% compared with a conventional word N-gram when applying the ATRASR engine. We also confirmed that our new model will be convenient for rapid development of an ASR system for resource-deficient languages, especially for agglutinative languages such as Mongolian.

UR - http://www.scopus.com/inward/record.url?scp=71249128067&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=71249128067&partnerID=8YFLogxK

U2 - 10.1109/ICSDA.2009.5278368

DO - 10.1109/ICSDA.2009.5278368

M3 - Conference contribution

AN - SCOPUS:71249128067

SN - 9781424444007

T3 - 2009 Oriental COCOSDA International Conference on Speech Database and Assessments, ICSDA 2009

SP - 104

EP - 109

BT - 2009 Oriental COCOSDA International Conference on Speech Database and Assessments, ICSDA 2009

T2 - 2009 Oriental COCOSDA International Conference on Speech Database and Assessments, ICSDA 2009

Y2 - 10 August 2009 through 12 August 2009

ER -

Modeling characteristics of agglutinative languages with multi-class language model for ASR system

Abstract

Publication series

Conference

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this