• Phonetics and Speech Sciences
• /
• v.15 no.1
• /
• pp.25-34
• /
• 2023

This study examined the utility of the acoustic features of vowels as cues for the place of articulation of Korean nasal consonants. In the acoustic analysis, spectral and temporal parameters were measured at the 25%, 50%, and 75% time points in the vowels neighboring nasal consonants in samples extracted from a spontaneous Korean speech corpus. Using these measurements, linear discriminant analyses were performed and classification accuracies for the nasal place of articulation were estimated. The analyses were applied separately for vowels following and preceding a nasal consonant to compare the effects of progressive and regressive coarticulation in terms of place of articulation. The classification accuracies ranged between approximately 50% and 60%, implying that acoustic measurements of vowel intervals alone are not sufficient to predict or classify the place of articulation of adjacent nasal consonants. However, given that these results were obtained for measurements at the temporal midpoint of vowels, where they are expected to be the least influenced by coarticulation, the present results also suggest the potential of utilizing acoustic measurements of vowels to improve the recognition accuracy of nasal place. Moreover, the classification accuracy for nasal place was higher for vowels preceding the nasal sounds, suggesting the possibility of higher anticipatory coarticulation reflecting the nasal place.

• Phonetics and Speech Sciences
• /
• v.14 no.1
• /
• pp.1-11
• /
• 2022

The purpose of this study is to investigate the difference of vowel duration due to the voicing of word-final consonants in English and its relation to the types of word-final consonants (stops vs. fricatives), (partial) devoicing, and stop releasing. Addtionally, this study attempts to interpret the findings from the functional view that the vowels before voiced consonants are produced with a longer duration in order to enhance the salience of the voicing of word-final consonants. This study conducted a recording experiment with English native speakers, and measured the vowel duration, the degree of (partial) devoicing of word-final voiced consonants and the release of word-final stops. First, the results showed that the ratio of the duration difference was not influenced by the types of word-final consonants. Second, it was revealed that the higher the degree of (partial) devoicing of word-final voiced consonants, the longer vowel duration before word-final voiced consonants, which was compatible with the prediction based on the functional view. Lastly, the ratio of the duration difference was greater when the word-final stops were uttered with the release compared to when uttered without the release, which was not consistent with the functional view. These results suggest that it is not sufficient enough to explain the voicing effect by its function of distinguishing the voicing of word-final consonants.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.4
• /
• pp.409-414
• /
• 2023

We described how to implement character expressions and animations that play an important role in expressing emotions and personalities in low-polygon character animation. With the development of the video industry, character expressions and mouth-shaped lip-syncing in animation can realize natural movements at a level close to real life. However, for non-experts, it is difficult to use expert-level advanced technology. Therefore, We aimed to present a guide for low-budget low-polygon character animators or non-experts to create mouth-shaped lip-syncing more naturally using accessible and highly usable features. A total of 8 mouth shapes were developed for mouth shape lip-sync animation: 'ㅏ', 'ㅔ', 'ㅣ', 'ㅗ', 'ㅜ', 'ㅡ', 'ㅓ' and a mouth shape that expresses a labial consonant. In the case of facial expression animation, a total of nine animations were produced by adding highly utilized interest, boredom, and pain to the six basic human emotions classified by Paul Ekman: surprise, fear, disgust, anger, happiness, and sadness. This study is meaningful in that it makes it easy to produce natural animation using the features built into the modeling program without using complex technologies or programs.

• Korean Journal of Cognitive Science
• /
• v.3 no.1
• /
• pp.61-78
• /
• 1991

This paper descibes the study of application of a modified Neocognitron model with backward path for the recognition of Hangul(Korean) syllabic characters. In this original report, Fukushima demonstrated that Neocognitron can recognize hand written numerical characters of $19{\times}19$ size. This version accepts $61{\times}61$ images of handwritten Hangul syllabic characters or a part thereof with a mouse or with a scanner. It consists of an input layer and 3 pairs of Uc layers. The last Uc layer of this version, recognition layer, consists of 24 planes of $5{\times}5$ cells which tell us the identity of a grapheme receiving attention at one time and its relative position in the input layer respectively. It has been trained 10 simple vowel graphemes and 14 simple consonant graphemes and their spatial features. Some patterns which are not easily trained have been trained more extrensively. The trained nerwork which can classify indivisual graphemes with possible deformation, noise, size variance, transformation or retation wre then used to recongnize Korean syllabic characters using its selective attention mechanism for image segmentation task within a syllabic characters. On initial sample tests on input characters our model could recognize correctly up to 79%of the various test patterns of handwritten Korean syllabic charactes. The results of this study indeed show Neocognitron as a powerful model to reconginze deformed handwritten charavters with big size characters set via segmenting its input images as recognizable parts. The same approach may be applied to the recogition of chinese characters, which are much complex both in its structures and its graphemes. But processing time appears to be the bottleneck before it can be implemented. Special hardware such as neural chip appear to be an essestial prerquisite for the practical use of the model. Further work is required before enabling the model to recognize Korean syllabic characters consisting of complex vowels and complex consonants. Correct recognition of the neighboring area between two simple graphemes would become more critical for this task.

• The Korean Journal of Air & Space Law and Policy
• /
• v.2
• /
• pp.31-67
• /
• 1990

One of the first issues which arose in the evolution of air law was the determination of the vertical limits of airspace over private property. In 1959 the UN in its Ad Hoc Committee on the Peaceful Uses of Outer Space, started to give attention to the question of the meaning of the term "outer space". Discussions in the United Nations regarding the delimitation issue were often divided between those in favour of a functional approach ("functionalists"), and those seeking the delineation of a boundary ("spatialists"). The functionalists, backed initially by both major space powers, which viewed any boundary as possibly restricting their access to space(Whether for peaceful or military purposes), won the first rounds, starting with the 1959 Report of the Ad Hoc Committee on the Peaceful Uses of Outer Space which did not consider that the topic called for priority consideration. In 1966, however, the spatialists, were able to place the issue on the agenda of the Outer Sapce Committee pursuant to Resolution 2222 (xxx1). However, the spatialists were not able to present a common position since there existed a variety of propositions for delineation of a boundary. Over the years, the funtionalists have seemed to be losing ground. As the element of location is a decisive factor for the choice of the legal regime to be applied, a purely functional approach to the regulation of activities in the space above the Earth does not offer a solution. It is therefore to be welcomed that there is clear evidence of a growing recognition of the defect inherent to such an approach and that a spatial approach to the problem is gaining support both by a growing number of States as well as by publicists. The search for a solution of the problem of demarcating the two different legal regimes governing the space above the Earth has undoubtedly been facilitated, and a number of countries, among them Argentina, Belgium, France, Italy and Mexico have already advocated the acceptance of the lower boundary of outer space at a height of 100km. The adoption of the principle of sovereignty at that height does not mean that States would not be allowed to take protective measures against space activities above that height which constitute a threat to their security. A parallel can be drawn with the defence of the State's security on the high seas. Measures taken by States in their own protection on the high seas outside the territorial waters-provided that they are proportionate to the danger-are not considered to infringe the principle of international law. The most important issue in this context relates to the problem of a right of passage for space craft through foreign air space in order to reach outer space. In the reports to former ILA Conferences an explanation was given of the reasons why no customary rule of freedom of passage for aircraft through foreign territorial air space could as yet be said to exist. It was suggested, however, that though the essential elements for the creation of a rule of customary international law allowing such passage were still lacking, developments apperaed to point to a steady growth of a feeling of necessity for such a rule. A definite treaty solution of the demarcation problem would require further study which should be carried out by the UN Outer Space Committee in close co-operation with other interested international organizations, including ICAO. If a limit between air space and outer space were established, air space would automatically come under the regime of the Chicago Convention alone. The use of the word "recognize" in Art. I of chicago convention is an acknowledgement of sovereignty over airspace existing as a general principle of law, the binding force of which exists independently of the Convention. Further it is important to note that the Aricle recognizes this sovereignty, as existing for every state, holding it immaterial whether the state is or is not a contracting state. The functional criteria having been created by reference to either the nature of activity or the nature of the space object, the next hurdle would be to provide methods of verification. With regard to the question of international verification the establishment of an International Satelite Monitoring Agency is required. The path towards the successful delimitation of outer space from territorial space is doubtless narrow and stony but the establishment of a precise legal framework, consonant with the basic principles of international law, for the future activities of states in outer space will, it is still believed, remove a source of potentially dangerous conflicts between states, and furthermore afford some safeguard of the rights and interests of non-space powers which otherwise are likely to be eroded by incipient customs based on at present almost complete freedom of action of the space powers.

• Journal of Intelligence and Information Systems
• /
• v.23 no.2
• /
• pp.71-88
• /
• 2017

Language models were originally developed for speech recognition and language processing. Using a set of example sentences, a language model predicts the next word or character based on sequential input data. N-gram models have been widely used but this model cannot model the correlation between the input units efficiently since it is a probabilistic model which are based on the frequency of each unit in the training set. Recently, as the deep learning algorithm has been developed, a recurrent neural network (RNN) model and a long short-term memory (LSTM) model have been widely used for the neural language model (Ahn, 2016; Kim et al., 2016; Lee et al., 2016). These models can reflect dependency between the objects that are entered sequentially into the model (Gers and Schmidhuber, 2001; Mikolov et al., 2010; Sundermeyer et al., 2012). In order to learning the neural language model, texts need to be decomposed into words or morphemes. Since, however, a training set of sentences includes a huge number of words or morphemes in general, the size of dictionary is very large and so it increases model complexity. In addition, word-level or morpheme-level models are able to generate vocabularies only which are contained in the training set. Furthermore, with highly morphological languages such as Turkish, Hungarian, Russian, Finnish or Korean, morpheme analyzers have more chance to cause errors in decomposition process (Lankinen et al., 2016). Therefore, this paper proposes a phoneme-level language model for Korean language based on LSTM models. A phoneme such as a vowel or a consonant is the smallest unit that comprises Korean texts. We construct the language model using three or four LSTM layers. Each model was trained using Stochastic Gradient Algorithm and more advanced optimization algorithms such as Adagrad, RMSprop, Adadelta, Adam, Adamax, and Nadam. Simulation study was done with Old Testament texts using a deep learning package Keras based the Theano. After pre-processing the texts, the dataset included 74 of unique characters including vowels, consonants, and punctuation marks. Then we constructed an input vector with 20 consecutive characters and an output with a following 21st character. Finally, total 1,023,411 sets of input-output vectors were included in the dataset and we divided them into training, validation, testsets with proportion 70:15:15. All the simulation were conducted on a system equipped with an Intel Xeon CPU (16 cores) and a NVIDIA GeForce GTX 1080 GPU. We compared the loss function evaluated for the validation set, the perplexity evaluated for the test set, and the time to be taken for training each model. As a result, all the optimization algorithms but the stochastic gradient algorithm showed similar validation loss and perplexity, which are clearly superior to those of the stochastic gradient algorithm. The stochastic gradient algorithm took the longest time to be trained for both 3- and 4-LSTM models. On average, the 4-LSTM layer model took 69% longer training time than the 3-LSTM layer model. However, the validation loss and perplexity were not improved significantly or became even worse for specific conditions. On the other hand, when comparing the automatically generated sentences, the 4-LSTM layer model tended to generate the sentences which are closer to the natural language than the 3-LSTM model. Although there were slight differences in the completeness of the generated sentences between the models, the sentence generation performance was quite satisfactory in any simulation conditions: they generated only legitimate Korean letters and the use of postposition and the conjugation of verbs were almost perfect in the sense of grammar. The results of this study are expected to be widely used for the processing of Korean language in the field of language processing and speech recognition, which are the basis of artificial intelligence systems.