• MALSORI
• /
• no.50
• /
• pp.139-150
• /
• 2004

A Phonetic Tied-Mixture (PTM) model has been proposed as a way of efficient decoding in large vocabulary continuous speech recognition systems (LVCSR). It has been reported that PTM model shows better performance in decoding than triphones by sharing a set of mixture components among states of the same topological location[5]. In this paper we propose a Phonetic Tied-Mixture Syllable (PTMS) model which extends PTM technique up to syllables. The proposed PTMS model shows 13% enhancement in decoding speed than PTM. In spite of difference in context dependent modeling (PTM : cross-word context dependent modeling, PTMS : word-internal left-phone dependent modeling), the proposed model shows just less than 1% degradation in word accuracy than PTM with the same beam width. With a different beam width, it shows better word accuracy than in PTM at the same or higher speed.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.33B no.10
• /
• pp.98-106
• /
• 1996

This paper proposes an algorithm for analyzing illocutionary force type (IfT)s in a dialogue, based on the context and modal information in the ending of a word. In korean, the variation of an illocutionary force type that represents a speaker's intention frequently occurs at the ending of a word, according to the type of modal information. And in an analysis of speech acts, the modal information illocutionary force types. In this paper, we analyze real dialogue dta, classify the types of illocutionary forces, perform the manual tagging of IFTs and show the freqency of each IFT's occurence. And we also propose an algorithm to extract IFTs, based on the relationship between the analyzed IFTs and the endings of a word. And we use this proposed algorithm to make an experiment on dialogue data and show its efficiency.

• Phonetics and Speech Sciences
• /
• v.3 no.1
• /
• pp.51-65
• /
• 2011

This study attempts to investigate the interlanguage speech intelligibility benefit for listeners (ISIB-L), examining Chinese talkers' production of English liquids and its perception of native listeners and non-native Chinese and Korean listeners. An Accent Judgment Task was conducted to measure non-native talkers' and listeners' phonological proficiency, and two levels of proficiency groups (high and low) participated in the experiment. The English liquids /l/ and /r/ produced by Chinese talkers were considered in terms of positions (syllable initial and final), contexts (segment, word and sentence) and lexical density (minimal vs. nonminimal pair) to see if these factors play a role in ISIIB-L. Results showed that both matched and mismatched interlanguage speech intelligibility benefit for listeners occurred except for the initial /l/. Non-native Chinese and Korean listeners, though only with high proficiency, were more accurate at identifying initial /r/, final /l/ and final /r/, but initial /l/ was significantly more intelligible to native listeners than non-native listeners. There was evidence of contextual and lexical density effects on ISIB-L. No ISIB-L was demonstrated in sentence context, but both matched and mismatched ISIB-L was observed in word context; this finding held true for only high proficiency listeners. Listeners recognized the targets better in the non-minimal pair (sparse density) environment than the minimal pair (higher density) environment. These findings suggest that ISIB-L for English liquids is influenced by talkers' and listeners' proficiency, syllable position in association with L1 and L2 phonological structure, context, and word neighborhood density.

• Korean Journal of Cognitive Science
• /
• v.27 no.1
• /
• pp.43-60
• /
• 2016

Recently, Unsupervised Word Sense Disambiguation research has focused on Graph based disambiguation. Graph-based disambiguation has built a semantic graph based on words collocated in context or sentence. However, building such a graph over all ambiguous word lead to unnecessary addition of edges and nodes (and hence increasing the error). In contrast, our work uses Word2Vec to consider the most similar words to an ambiguous word in the context or sentences, to rebuild a graph of the matched words. As a result, we show a higher F1-Measure value than the previous methods by using Word2Vec.

• Journal of Computing Science and Engineering
• /
• v.11 no.1
• /
• pp.32-38
• /
• 2017

Determining the correct word sense among ambiguous senses is essential for semantic analysis. One of the models for word sense disambiguation is the word space model which is very simple in the structure and effective. However, when the context word vectors in the word space model are merged into sense vectors in a sense inventory, they become typically very large but still suffer from the lexical scarcity. In this paper, we propose a word sense disambiguation method using word embedding that makes the sense inventory vectors compact and efficient due to its additive compositionality. Results of experiments with a Korean sense-tagged corpus show that our method is very effective.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.43 no.6 s.312
• /
• pp.43-51
• /
• 2006

This paper proposes a metrics for example matching under the example-based machine translation for English-Korean machine translation. Our metrics served as similarity measure is based on edit-distance algorithm, and it is employed to retrieve the most similar example sentences to a given query. Basically it makes use of simple information such as lemma and part-of-speech information of typographically mismatched words. Edit-distance algorithm cannot fully reflect the context of matched word units. In other words, only if matched word units are ordered, it is considered that the contribution of full matching context to similarity is identical to that of partial matching context for the sequence of words in which mismatching word units are intervened. To overcome this drawback, we propose the context-weighting scheme that uses the contiguity information of matched word units to catch the full context. To change the edit-distance metrics representing dissimilarity to similarity metrics, to apply this context-weighted metrics to the example matching problem and also to rank by similarity, we normalize it. In addition, we generalize previous methods using some linguistic information to one representative system. In order to verify the correctness of the proposed context-weighted metrics, we carry out the experiment to compare it with generalized previous methods.

• The Korean Journal of Applied Statistics
• /
• v.34 no.3
• /
• pp.389-399
• /
• 2021

In recent years, word embedding has been a popular field of natural language processing research and a skip-gram has become one successful word embedding method. It assigns a word embedding vector to each word using contexts, which provides an effective way to analyze text data. However, due to the limitation of vector space model, primary word embedding methods assume that every word only have a single meaning. As one faces multi-sense words, that is, words with more than one meaning, in reality, Neelakantan (2014) proposed a multi-sense skip-gram (MSSG) to find embedding vectors corresponding to the each senses of a multi-sense word using a clustering method. In this paper, we propose a modified method of the MSSG to improve statistical accuracy. Moreover, we propose a data-adaptive choice of the number of clusters, that is, the number of meanings for a multi-sense word. Some numerical evidence is given by conducting real data-based simulations.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.24 no.5
• /
• pp.470-476
• /
• 2014

A point of interest is a specific point location such as a cafe, a gallery, a shop, or a park. It consists of a name, a category, a location, and so on. Its information is necessary for location-based application, above all category is basic information. However, category information should be automatically gathered because it costs high to gather it manually. In this paper, we propose a novel method to estimate category of POIs automatically using an inner word and local context. An inner word is a word that contains POI's name. Their name sometimes expose category information. Thus, their name is used as inner word information in estimating category of POIs. Local context information means words around a POI's name in a document that mentioned the name. The context include information to estimate category. The evaluation of the proposed method is performed on two data sets. According to the experimental results, proposed model using combination inner word and local context show higher accuracy than that of model using each.

• ETRI Journal
• /
• v.35 no.1
• /
• pp.89-99
• /
• 2013

The problem of the sentence-based pronunciation evaluation task is defined in the context of subjective criteria. Three subjective criteria (that is, the minimum subjective word score, the mean subjective word score, and first impression) are proposed and modeled with the combination of word-based assessment. Then, the subjective criteria are approximated with objective sentence pronunciation scores obtained with the combination of word-based metrics. No a priori studies of common mistakes are required, and class-based language models are used to incorporate incorrect and correct pronunciations. Incorrect pronunciations are automatically incorporated by making use of a competitive lexicon and the phonetic rules of students' mother and target languages. This procedure is applicable to any second language learning context, and subjective-objective sentence score correlations greater than or equal to 0.5 can be achieved when the proposed sentence-based pronunciation criteria are approximated with combinations of word-based scores. Finally, the subjective-objective sentence score correlations reported here are very comparable with those published elsewhere resulting from methods that require a priori studies of pronunciation errors.

• Journal of Science Education
• /
• v.46 no.1
• /
• pp.109-120
• /
• 2022

Fraction division is the content that is important but difficult to learn because it includes the process of finding a numerical expression in the real-world context, the process of making a context that matches a numerical expression, how to solve division, and the justification of standard algorithm. This study analyzes the word problems and problem solving methods about fraction division which elementary pre-service teachers represented. Pre-service teachers have more difficulty in making word problem where the dividend is less than the divisor and they also show typical errors in making the word problems. There were differences in the methods presented according to the contexts of division in problem solving. Through this study, it is necessary to rethink the teaching methods for fraction division instruction in the curriculum for pre-service teachers and analyze the formation process of 'knowledge for content and teaching' because of the differences in responses between grades.