• Information and Communications Magazine
• /
• v.7 no.3
• /
• pp.16-26
• /
• 1990

본 논문은 음성인식을 위한 neural network의 하나인 TRACE model에 대하여 기술한다. TRACE model은 HEARSAY 음성이해 system의 blackboard 구조와 유사한 trace 구조를 사용하여 정보처리 unit들간의 자극성 내지 억제성 연결에 의한 상호작용의 결과로서 음성을 인식한다. TRACE model은 특징층, 음소층 그리고 단어층의 3층으로 구성되며 층간의 unit들이 상호작용할 뿐 아니라 동일층의 unit들이 상호경쟁함으로써 음성인식에 있어서의 context 효과, segmentation 및 잡음 등의 문제를 해결할 수 있다.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.4 no.2
• /
• pp.116-122
• /
• 2009

In this paper, we propose a new method which generates the trajectory of the mouth shape for the characters by the user inputs. It is based on the character at a basis syllable and can be suitable to the mouth shape generation. In this paper, we understand the principle of the Korean language creation and find the similarity for the form of the mouth shape and select it as a basic syllable. We also consider the articulation of this phoneme for it and create a new mouth shape trajectory and apply at face of an 3D avatar.

• The Journal of the Acoustical Society of Korea
• /
• v.22 no.5
• /
• pp.388-402
• /
• 2003

In this paper, we carried out the word, 4 continuous digits. continuous, and task-independent word recognition experiments to verify the effectiveness of the re-defined phoneme-likely units (PLUs) for the phonetic decision tree based HM-Net (Hidden Markov Network) context-dependent (CD) acoustic modeling in Korean appropriately. In case of the 48 PLUs, the phonemes /ㅂ/, /ㄷ/, /ㄱ/ are separated by initial sound, medial vowel, final consonant, and the consonants /ㄹ/, /ㅈ/, /ㅎ/ are also separated by initial sound, final consonant according to the position of syllable, word, and sentence, respectively. In this paper. therefore, we re-define the 39 PLUs by unifying the one phoneme in the separated initial sound, medial vowel, and final consonant of the 48 PLUs to construct the CD acoustic models effectively. Through the experimental results using the re-defined 39 PLUs, in word recognition experiments with the context-independent (CI) acoustic models, the 48 PLUs has an average of 7.06%, higher recognition accuracy than the 39 PLUs used. But in the speaker-independent word recognition experiments with the CD acoustic models, the 39 PLUs has an average of 0.61% better recognition accuracy than the 48 PLUs used. In the 4 continuous digits recognition experiments with the liaison phenomena. the 39 PLUs has also an average of 6.55% higher recognition accuracy. And then, in continuous speech recognition experiments, the 39 PLUs has an average of 15.08% better recognition accuracy than the 48 PLUs used too. Finally, though the 48, 39 PLUs have the lower recognition accuracy, the 39 PLUs has an average of 1.17% higher recognition characteristic than the 48 PLUs used in the task-independent word recognition experiments according to the unknown contextual factor. Through the above experiments, we verified the effectiveness of the re-defined 39 PLUs compared to the 48PLUs to construct the CD acoustic models in this paper.

• Korean Journal of Cognitive Science
• /
• v.19 no.2
• /
• pp.163-175
• /
• 2008

Twelve Japanese students living in South Korea have been examined for their perceptual identification of an initial consonant in Korean syllables with or without a white noise. A confusion matrix was then subject to analyses of additive clustering, individual difference scaling, and probability of information transmission, the results of which were also compared to those of South Koreans. The Japanese in the present experiment confused /다/and/타/ most frequently, followed by /가/ and /카/, /자, 차, 짜/, /타/ and /따/, and so on. The results of additive clustering analysis of the Japanese significantly differed from those of the South Koreans. Individual difference scaling revealed dimensions of sonorant, aspiration and coronal. While South Koreans showed binary values on aspiration and tenseness dimensions, the Japanese did continuous values on such dimensions. An information transmission probability analysis revealed that the Japanese participants could not perceive very well such larynx features as tenseness and aspiration compared to the South Korean participants. The former group, however, perceived very well place of articulation features such as labial and coronal. The present results suggest that an approach dealing with structures of base representations is important in understanding the phonological categories of languages.

• Korean Journal of Cognitive Science
• /
• v.26 no.3
• /
• pp.263-278
• /
• 2015

Online Q&A for the National Institute of the Korean Language provides expert's answers for questions about the Korean language, in which many similar questions are repeatedly posted like other Q&A boards. So, if a system automatically finds questions that are similar to a user's question, it can immediately provide users with recommendable answers to their question and prevent experts from wasting time to answer to similar questions repeatedly. In this paper, we set 5 classes of questions based on its topic which are frequently asked, and propose to classify questions to those classes. Our system searches similar questions by combining topic similarity, vector similarity and sequence similarity. Experiment shows that our method improves search correctness with topic classification. In experiment, Mean Reciprocal Rank(MRR) of our system is 0.756, and precision for the first result is 68.31% and precision for top five results is 87.32%.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.21 no.3
• /
• pp.600-611
• /
• 1996

Recently, neural network-based speech recognition has been studied to utilize the adaptivity and learnability of neural network models. However, conventional neural network models have difficulty in the co-articulation processing and the boundary detection of similar phonmes of the Korean speech. Also, in case of using one phonotopic map, learning speed may dramatically increase and inaccuracies may be caused because homogeneous learning and recognition method should be applied for heterogenous data. Hence, in this paper, a neural net typewriter has been designed using a hierarchical self-organizing classifier(HSOC), and related algorithms are presented. This HSOC, during its learing stage, distributed phoneme data on hierarchically structured multiple phonotopic maps, using Kohonen's self-organizing feature maps(SOFM). Presented and experimented in this paper were the algorithms for deciding the number of maps, map sizes, the selection of phonemes and their placement per map, an approapriate learning and preprocessing method per map. If maps are divided according to a priorlinguistic knowledge, we would have difficulty in acquiring linguistic knowledge and how to alpply it(e.g., processing extended phonemes). Contrarily, our HSOC has an advantage that multiple phonotopic maps suitable for given input data are self-organizable. The resulting three korean phonotopic maps are optimally labelled and have their own optimal preprocessing schemes, and also confirm to the conventional linguistic knowledge.

• The Journal of the Acoustical Society of Korea
• /
• v.20 no.2
• /
• pp.27-34
• /
• 2001

Utterance verification is used in variable vocabulary word recognition to reject the word that does not belong to in-vocabulary word or does not belong to correctly recognized word. Utterance verification is an important technology to design a user-friendly speech recognition system. We propose a new utterance verification algorithm for no-training utterance verification system based on the minimum verification error. First, using PBW (Phonetically Balanced Words) DB (445 words), we create no-training anti-phoneme models which include many PLUs(Phoneme Like Units), so anti-phoneme models have the minimum verification error. Then, for OOV (Out-Of-Vocabulary) rejection, the phoneme-based confidence measure which uses the likelihood between phoneme model (null hypothesis) and anti-phoneme model (alternative hypothesis) is normalized by null hypothesis, so the phoneme-based confidence measure tends to be more robust to OOV rejection. And, the word-based confidence measure which uses the phoneme-based confidence measure has been shown to provide improved detection of near-misses in speech recognition as well as better discrimination between in-vocabularys and OOVs. Using our proposed anti-model and confidence measure, we achieve significant performance improvement; CA (Correctly Accept for In-Vocabulary) is about 89％, and CR (Correctly Reject for OOV) is about 90％, improving about 15-21％ in ERR (Error Reduction Rate).

• The Journal of the Acoustical Society of Korea
• /
• v.28 no.6
• /
• pp.572-577
• /
• 2009

Large corpus-based concatenating Text-to-Speech (TTS) systems can generate natural synthetic speech without additional signal processing. To prune the redundant speech segments in a large speech segment DB, we can utilize a decision-tree based triphone clustering algorithm widely used in speech recognition area. But, the conventional methods have problems in representing the acoustic transitional characteristics of the phones and in applying context questions with hierarchic priority. In this paper, we propose a new clustering algorithm to downsize the speech DB. Firstly, three 13th order MFCC vectors from first, medial, and final frame of a phone are combined into a 39 dimensional vector to represent the transitional characteristics of a phone. And then the hierarchically grouped three question sets are used to construct the triphone trees. For the performance test, we used DTW algorithm to calculate the acoustic similarity between the target triphone and the triphone from the tree search result. Experimental results show that the proposed method can reduce the size of speech DB by 23% and select better phones with higher acoustic similarity. Therefore the proposed method can be applied to make a small sized TTS.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.18 no.6
• /
• pp.241-250
• /
• 2019

Road signs are guide facilities for road users, and the Ministry of Land, Infrastructure and Transport has established and operated a system to enhance the convenience of managing these road signs. The role of road signs will decrease in the future autonomous driving, but they will continue to be needed. For the accurate mechanical recognition of texts on road signs, automatic road sign recognition equipment has been developed and it has applied image-based text recognition technology. Yet there are many cases of misrecognition due to irregular specifications and external environmental factors such as manual manufacturing, illumination, light reflection, and rainfall. The purpose of this study is to derive location-based destination names for finding misrecognition errors that cannot be overcome by image analysis, and to improve the automatic recognition of road signs destination names by using Levenshtein similarity verification method based on phoneme separation.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.7 no.2
• /
• pp.177-182
• /
• 2003

In this study, we research the robust speech recognition for the syllables and phoneme units with the feature parameter including the voiced-unvoiced measures for the confusable words. In order to make it possible, we propose the measure representing the voiced-unvoiced degree by using the HPS(Harmonic Product Spectrum) information, used on pitch detection. We proposed this measures with the sharpnes, peak count and height measure of HPS. We reconstructed the feature parameter including this measures, then we performs the speech recognition experiments and compared with the typical feature parameters under the CVC type confusable syllables DB.