Acoustic model adaptation based on pronunciation variability analysis for non-native speech recognition

Oh, Yoo Rhee; Yoon, Jae Sam; Kim, Hong Kook

doi:10.1016/j.specom.2006.10.006

Cited by 38 publications

(32 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In other words, the acoustic models trained with two different languages are combined to obtain the acoustic models for non-native speech. However, the most popular way of obtaining the adjusted acoustic models is to apply an adaptation technique with only small amount of adaptation data for non-native speech (Liu et al, 2008;Oh et al, 2007;). …”

Section: Overview Of Non-native Speech Recognitionmentioning

confidence: 99%

“…In other words, acoustic or pronunciation modeling approaches can be combined in an MLLR and/or MAP adaptation framework (Goronzy et al, 2004;He et al, 2003;Liu et al, 2008;Oh et al, 2007;Tan et al, 2007). In particular, Bouselmi et al (Bouselmi et al, 2007) proposed several combination schemes for pronunciation and MLLR/MAP acoustic model adaptations.…”

Section: Hybrid Modeling Approachmentioning

confidence: 99%

“…Contrary to interpolating acoustic models, phone acoustic models of the target language were modified by adding an alternative path to the corresponding mother tongue phone acoustic models of non-native speakers (Bartkova et al, 2006;Bouselmi et al, 2006). Finally, the acoustic models were adapted by using non-native adaptation data via either an algorithm dedicated to nonnative ASR or a conventional speaker adaptation method (Liu et al, 2008;Oh et al, 2007;.…”

Section: Acoustic Modeling Approachmentioning

confidence: 99%

See 2 more Smart Citations

Non-Native Pronunciation Variation Modeling for Automatic Speech Recognition

Kim¹,

Oh²,

Kim³

2010

Advances in Speech Recognition

View full text Add to dashboard Cite

Section: Overview Of Non-native Speech Recognitionmentioning

confidence: 99%

Section: Hybrid Modeling Approachmentioning

confidence: 99%

Section: Acoustic Modeling Approachmentioning

confidence: 99%

See 1 more Smart Citation

Non-Native Pronunciation Variation Modeling for Automatic Speech Recognition

Kim¹,

Oh²,

Kim³

2010

Advances in Speech Recognition

View full text Add to dashboard Cite

“…The vocabulary and grammar of nonnative speakers is often limited and therefore basic, but a speech recognizer takes no or only a little advantage of this and is confused by the different phonetics [1]. In order to improve the speech recognition accuracy for non-native speech, various methodologies have been proposed for adapting to the acoustic features of non-native speech, including a speaker adaptation method for second language speech recognition [2], a method using the state-tying of acoustic modeling (AM) for second language speech with a variant phonetic unit obtained by analyzing the variability of second language speech pronunciation [3], AM interpolating with both native and non-native acoustic models [4], and others. Automatic speech recognition (ASR) technology for non-native speech adopted in various speech dialogue systems was developed assuming that the mother tongues of users were both unknown and various.…”

Section: Introductionmentioning

confidence: 99%

Phoneme Set Design for Speech Recognition of English by Japanese

Wang

Zhang

Nishida

et al. 2015

IEICE Trans. Inf. ^|^ Syst.

View full text Add to dashboard Cite

SUMMARYThis paper describes a novel method to improve the performance of second language speech recognition when the mother tongue of users is known. Considering that second language speech usually includes less fluent pronunciation and more frequent pronunciation mistakes, the authors propose using a reduced phoneme set generated by a phonetic decision tree (PDT)-based top-down sequential splitting method instead of the canonical one of the second language. The authors verify the efficacy of the proposed method using second language speech collected with a translation game type dialogue-based English CALL system. Experiments show that a speech recognizer achieved higher recognition accuracy with the reduced phoneme set than with the canonical phoneme set. key words: phonetic decision tree (PDT), Phoneme set, Second language speech recognition

show abstract

“…Non-native speech poses several challenges for automatic speech recognition. In order to improve the speech recognition accuracy for non-native speech, various methodologies have been proposed, including acoustic model adaptation for second language speech with a variant phonetic unit obtained by analyzing the variability of second language speech pronunciation [2], an acoustic model interpolating from both native and non-native acoustic models [3], data driven generation of pronunciation variants for lexical modeling [4], and others. These automatic speech recognition technologies for non-native speech have been developed assuming the mother tongues of users to be unknown and various.…”

Section: Introductionmentioning

confidence: 99%

Speech Recognition of English by Japanese Using Lexicon Represented by Multiple Reduced Phoneme Sets

Wang

Yamamoto

2015

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

SUMMARYRecognition of second language (L2) speech is still a challenging task even for state-of-the-art automatic speech recognition (ASR) systems, partly because pronunciation by L2 speakers is usually significantly influenced by the mother tongue of the speakers. The authors previously proposed using a reduced phoneme set (RPS) instead of the canonical one of L2 when the mother tongue of speakers is known, and demonstrated that this reduced phoneme set improved the recognition performance through experiments using English utterances spoken by Japanese. However, the proficiency of L2 speakers varies widely, as does the influence of the mother tongue on their pronunciation. As a result, the effect of the reduced phoneme set is different depending on the speakers' proficiency in L2. In this paper, the authors examine the relation between proficiency of speakers and a reduced phoneme set customized for them. The experimental results are then used as the basis of a novel speech recognition method using a lexicon in which the pronunciation of each lexical item is represented by multiple reduced phoneme sets, and the implementation of a language model most suitable for that lexicon is described. Experimental results demonstrate the high validity of the proposed method. key words: second language (L2) speech recognition, proficiency of L2 speakers, reduced phoneme set (RPS), multiple reduced phoneme sets, proficiency-dependent reduced phoneme set

show abstract

Acoustic model adaptation based on pronunciation variability analysis for non-native speech recognition

Cited by 38 publications

References 11 publications

Non-Native Pronunciation Variation Modeling for Automatic Speech Recognition

Non-Native Pronunciation Variation Modeling for Automatic Speech Recognition

Phoneme Set Design for Speech Recognition of English by Japanese

Speech Recognition of English by Japanese Using Lexicon Represented by Multiple Reduced Phoneme Sets

Contact Info

Product

Resources

About