• Korean Journal of Cognitive Science
• /
• v.26 no.3
• /
• pp.323-355
• /
• 2015

This study examined grapheme-to-phoneme regularity effect among late Korean-English bilinguals by using whole word level task (lexical processing) and two meta-phonological tasks(sub-lexical processing): [1] English word naming task(whole word level), [2] rhyme judgement task(rhyme level), and [3] phoneme deletion task(phoneme level). Forty-three late Korean-English bilinguals participated in all three tasks. In these tasks, participants showed better performance in regular word conditions compared to irregular word conditions, demonstrating a clear English regularity effect. Post-hoc correlational analysis revealed strong correlation between word naming task and rhyme judgement task, which is different from the results reported with English monolinguals. The contradicting results might be due to the relevantly low English proficiency level among late Korean-English bilingual speakers. In conclusion, this study suggests that late Korean-English bilinguals make use of L2 grapheme-to-phoneme conversion (GPC) rule when reading L2 English words.

• Proceedings of the Korean Society for Cognitive Science Conference
• /
• 2000.05a
• /
• pp.292-296
• /
• 2000

뇌손상에 기인하는 한글 난독증의 어휘처리 양상을 분석하여 한글정보처리의 특성을 알아보고자 하였다. 피험자 PSK의 한글 어휘처리에서 특히 주목되는 점은 단어의 음독은 가능하나, 비단어의 음독은 불가능하였다는 것이다. PSK의 한글 어휘처리는, 자소-음소변환(grapheme-phoneme conversion)경로가 선택적으로 손상되어, 심성어휘집(mental lexicon)의 발음정보를 이용하는 직접경로에 의해서 이루어진다고 판단된다. 읽기(reading)와 그림명명(picture naming)에서 나타난 오류들에 대하여, 음운적 단서(phonological cueing)를 제시하였다. 그 결과, 읽기 수행에서는 단서 주기 효과가 나타나지 않았으나 그림명명에서는 수행상의 향상이 나타났다. 또한, 1음절어의 읽기 수행에서는 규칙효과가 나타나지 않았으나 2음절어의 읽기 수행에서는 빈도와 규칙성의 상호작용이 나타났다. 이것은, PSK의 1음절어와 2음절어에 대한 읽기 수행이 상이한 경로에서 이루어질 가능성을 시사한다.

• The Journal of the Acoustical Society of Korea
• /
• v.18 no.5
• /
• pp.21-27
• /
• 1999

In this paper, we propose an improved MLP-based English pronunciation dictionary generator to apply to the variable vocabulary word recognizer. The variable vocabulary word recognizer can process any words specified in Korean word lexicon dynamically determined according to the current recognition task. To extend the ability of the system to task for English words, it is necessary to build a pronunciation dictionary generator to be able to process words not included in a predefined lexicon, such as proper nouns. In order to build the English pronunciation dictionary generator, we use context-dependent grapheme-to-phoneme multi-layer perceptron(MLP) architecture for each grapheme. To train each MLP, it is necessary to obtain grapheme-to-phoneme training data from general pronunciation dictionary. To automate the process, we use dynamic programming(DP) algorithm with some distance metrics. For training and testing the grapheme-to-phoneme MLPs, we use general English pronunciation dictionary with about 110 thousand words. With 26 MLPs each having 30 to 50 hidden nodes and the exception grapheme lexicon, we obtained the word accuracy of 72.8% for the 110 thousand words superior to rule-based method showing the word accuracy of 24.0%.

• The Journal of the Acoustical Society of Korea
• /
• v.40 no.5
• /
• pp.530-536
• /
• 2021

Recently, due to the development of deep learning, end-to-end speech recognition, which directly maps graphemes to speech signals, shows good performance. Especially, among the end-to-end models, conformer shows the best performance. However end-to-end models only focuses on the probability of which grapheme will appear at the time. The decoding process uses a greedy search or beam search. This decoding method is easily affected by the final probability output by the model. In addition, the end-to-end models cannot use external pronunciation and language information due to structual problem. Therefore, in this paper conformer with lexicon transducer is proposed. We compare phoneme-based model with lexicon transducer and grapheme-based model with beam search. Test set is consist of words that do not appear in training data. The grapheme-based conformer with beam search shows 3.8 % of CER. The phoneme-based conformer with lexicon transducer shows 3.4 % of CER.