@inproceedings{talley-2006-bootstrapping,
title = "Bootstrapping New Language {ASR} Capabilities: Achieving Best Letter-to-Sound Performance under Resource Constraints",
author = "Talley, Jim",
editor = "Calzolari, Nicoletta and
Choukri, Khalid and
Gangemi, Aldo and
Maegaard, Bente and
Mariani, Joseph and
Odijk, Jan and
Tapias, Daniel",
booktitle = "Proceedings of the Fifth International Conference on Language Resources and Evaluation ({LREC}{'}06)",
month = may,
year = "2006",
address = "Genoa, Italy",
publisher = "European Language Resources Association (ELRA)",
url = "http://www.lrec-conf.org/proceedings/lrec2006/pdf/436_pdf.pdf",
abstract = "One of the most critical components in the process of building automatic speech recognition (ASR) capabilities for a new language is the lexicon, or pronouncing dictionary. For practical reasons, it is desirable to manually create only the minimal lexicon using available native-speaker phonetic expertise and, then, use the resulting seed lexicon for machine learning based induction of a high-quality letter-to-sound (L2S) model for generation of pronunciations for the remaining words of the language. This paper examines the viability of this scenario, specifically investigating three possible strategies for selection of lexemes (words) for manual transcription choosing the most frequent lexemes of the language, choosing lexemes randomly, and selection of lexemes via an information theoretic diversity measure. The relative effectiveness of these three strategies is evaluated as a function of the number of lexemes to be transcribed to create a bootstrapping lexicon. Generally, the newly developed orthographic diversity based selection strategy outperforms the others for this scenario where a limited number of lexemes can be transcribed. The experiments also provide generally useful insight into expected L2S accuracy sacrifice as a function of decreasing training set size.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="talley-2006-bootstrapping">
<titleInfo>
<title>Bootstrapping New Language ASR Capabilities: Achieving Best Letter-to-Sound Performance under Resource Constraints</title>
</titleInfo>
<name type="personal">
<namePart type="given">Jim</namePart>
<namePart type="family">Talley</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2006-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Fifth International Conference on Language Resources and Evaluation (LREC’06)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Nicoletta</namePart>
<namePart type="family">Calzolari</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Khalid</namePart>
<namePart type="family">Choukri</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aldo</namePart>
<namePart type="family">Gangemi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bente</namePart>
<namePart type="family">Maegaard</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Joseph</namePart>
<namePart type="family">Mariani</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jan</namePart>
<namePart type="family">Odijk</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Daniel</namePart>
<namePart type="family">Tapias</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>European Language Resources Association (ELRA)</publisher>
<place>
<placeTerm type="text">Genoa, Italy</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>One of the most critical components in the process of building automatic speech recognition (ASR) capabilities for a new language is the lexicon, or pronouncing dictionary. For practical reasons, it is desirable to manually create only the minimal lexicon using available native-speaker phonetic expertise and, then, use the resulting seed lexicon for machine learning based induction of a high-quality letter-to-sound (L2S) model for generation of pronunciations for the remaining words of the language. This paper examines the viability of this scenario, specifically investigating three possible strategies for selection of lexemes (words) for manual transcription choosing the most frequent lexemes of the language, choosing lexemes randomly, and selection of lexemes via an information theoretic diversity measure. The relative effectiveness of these three strategies is evaluated as a function of the number of lexemes to be transcribed to create a bootstrapping lexicon. Generally, the newly developed orthographic diversity based selection strategy outperforms the others for this scenario where a limited number of lexemes can be transcribed. The experiments also provide generally useful insight into expected L2S accuracy sacrifice as a function of decreasing training set size.</abstract>
<identifier type="citekey">talley-2006-bootstrapping</identifier>
<location>
<url>http://www.lrec-conf.org/proceedings/lrec2006/pdf/436_pdf.pdf</url>
</location>
<part>
<date>2006-05</date>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Bootstrapping New Language ASR Capabilities: Achieving Best Letter-to-Sound Performance under Resource Constraints
%A Talley, Jim
%Y Calzolari, Nicoletta
%Y Choukri, Khalid
%Y Gangemi, Aldo
%Y Maegaard, Bente
%Y Mariani, Joseph
%Y Odijk, Jan
%Y Tapias, Daniel
%S Proceedings of the Fifth International Conference on Language Resources and Evaluation (LREC’06)
%D 2006
%8 May
%I European Language Resources Association (ELRA)
%C Genoa, Italy
%F talley-2006-bootstrapping
%X One of the most critical components in the process of building automatic speech recognition (ASR) capabilities for a new language is the lexicon, or pronouncing dictionary. For practical reasons, it is desirable to manually create only the minimal lexicon using available native-speaker phonetic expertise and, then, use the resulting seed lexicon for machine learning based induction of a high-quality letter-to-sound (L2S) model for generation of pronunciations for the remaining words of the language. This paper examines the viability of this scenario, specifically investigating three possible strategies for selection of lexemes (words) for manual transcription choosing the most frequent lexemes of the language, choosing lexemes randomly, and selection of lexemes via an information theoretic diversity measure. The relative effectiveness of these three strategies is evaluated as a function of the number of lexemes to be transcribed to create a bootstrapping lexicon. Generally, the newly developed orthographic diversity based selection strategy outperforms the others for this scenario where a limited number of lexemes can be transcribed. The experiments also provide generally useful insight into expected L2S accuracy sacrifice as a function of decreasing training set size.
%U http://www.lrec-conf.org/proceedings/lrec2006/pdf/436_pdf.pdf
Markdown (Informal)
[Bootstrapping New Language ASR Capabilities: Achieving Best Letter-to-Sound Performance under Resource Constraints](http://www.lrec-conf.org/proceedings/lrec2006/pdf/436_pdf.pdf) (Talley, LREC 2006)
ACL