• Journal of the Institute of Convergence Signal Processing
• /
• v.15 no.2
• /
• pp.37-41
• /
• 2014

The vector Taylor series (VTS) based method usually employs clean speech Hidden Markov Models (HMMs) when compensating speech feature vectors or adapting the parameters of trained HMMs. It is well-known that noisy speech HMMs trained by the Multi-condition TRaining (MTR) and the Multi-Model-based Speech Recognition framework (MMSR) method perform better than the clean speech HMM in noisy speech recognition. In this paper, we propose a method to use the noise-adapted HMMs in the VTS-based speech feature compensation method. We derived a novel mathematical relation between the train and the test noisy speech feature vector in the log-spectrum domain and the VTS is used to estimate the statistics of the test noisy speech. An iterative EM algorithm is used to estimate train noisy speech from the test noisy speech along with noise parameters. The proposed method was applied to the noise-adapted HMMs trained by the MTR and MMSR and could reduce the relative word error rate significantly in the noisy speech recognition experiments on the Aurora 2 database.

• Journal of Information Processing Systems
• /
• v.13 no.4
• /
• pp.863-875
• /
• 2017

The traditional text similarity measurement methods based on word frequency vector ignore the semantic relationships between words, which has become the obstacle to text similarity calculation, together with the high-dimensionality and sparsity of document vector. To address the problems, the improved singular value decomposition is used to reduce dimensionality and remove noises of the text representation model. The optimal number of singular values is analyzed and the semantic relevance between words can be calculated in constructed semantic space. An inverted index construction algorithm and the similarity definitions between vectors are proposed to calculate the similarity between two documents on the semantic level. The experimental results on benchmark corpus demonstrate that the proposed method promotes the evaluation metrics of F-measure.

• Journal of information and communication convergence engineering
• /
• v.21 no.1
• /
• pp.9-16
• /
• 2023

The natural language on social network platforms has a certain front-to-back dependency in structure, and the direct conversion of Chinese text into a vector makes the dimensionality very high, thereby resulting in the low accuracy of existing text classification methods. To this end, this study establishes a deep learning model that combines a big data ultra-deep convolutional neural network (UDCNN) and long short-term memory network (LSTM). The deep structure of UDCNN is used to extract the features of text vector classification. The LSTM stores historical information to extract the context dependency of long texts, and word embedding is introduced to convert the text into low-dimensional vectors. Experiments are conducted on the social network platforms Sogou corpus and the University HowNet Chinese corpus. The research results show that compared with CNN + rand, LSTM, and other models, the neural network deep learning hybrid model can effectively improve the accuracy of text classification.

• The Journal of the Acoustical Society of Korea
• /
• v.5 no.3
• /
• pp.50-57
• /
• 1986

유한상태 벡터양자화 방법을 이용한 격리단어인식에 관하여 기술하고 있다. 이 인식시스템은 codebook과 next-state function 으로 구성된 일종의 finite-state machine으로 볼 수 있다. 유한상태 벡 터양자화방법을 이용한 격리단어 인식시스템은 일반적인 벡터양자화방법을 이용한 인식시스템에 비하여 소요시간이 감소하며 입력음성을 분할할 필요도 없는 한편 두 시스템의 인식율은 비슷한 것으로 나타났 다. Next-state function을 구하는 방법에는 conditional histogram 방법과 omniscient design 방법이 있 으며, 이 방법들의 성능비교를 위해 영부터 구까지의 한국어 숫자음성에 대한 인식실험을 수행하였다.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.36 no.3
• /
• pp.185-196
• /
• 2018

In order to overcome the problem that outdoor and indoor spatial information service are separately utilized, an integration service system of spatial information that is linked from outdoor to indoor has been implemented. As a result of the study, "0001.xml" corresponding to the file index key value, which is the service connection information in the building information of the destination, was extracted from the prototype verification of the system, the search word of 'Kim AB' was transmitted to the indoor map server and converted from the outdoor map service to the indoor map service through confirmation of the navigation service connected information, using service linkage information and search words of the indoor map service was confirmed that the route was displayed from the entrance of the building to the destination in the building through the linkage search DB (Database) table and the search query. Therefore, through this study was examined the possibility of linking indoor and outdoor DB through vector spatial information integration service system. The indoor map and the map engine were implemented based on the same vector map format as the outdoor map engine, it was confirmed that the connectivity of the map engine can be applied.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.7
• /
• pp.2785-2799
• /
• 2020

We propose a novel speaker adaptation method using acoustic model clustering. The similarity of different speakers is defined by the cosine distance between their i-vectors (intermediate vectors), and various efficient clustering algorithms are applied to obtain a number of speaker subsets with different characteristics. The speaker-independent model is then retrained with the training data of the individual speaker subsets grouped by the clustering results, and an unknown speech is recognized by the retrained model of the closest cluster. The proposed method is applied to a large-scale speech recognition system implemented by a hybrid hidden Markov model and deep neural network framework. An experiment was conducted to evaluate the word error rates using Resource Management database. When the proposed speaker adaptation method using i-vector based clustering was applied, the performance, as compared to that of the conventional speaker-independent speech recognition model, was improved relatively by as much as 12.2% for the conventional fully neural network, and by as much as 10.5% for the bidirectional long short-term memory.

• The KIPS Transactions:PartB
• /
• v.17B no.3
• /
• pp.249-254
• /
• 2010

We propose an automatic spam filter for e-mail data using Support Vector Machines(SVM). We use a lexical form of a word and its part of speech(POS) tags as features and select features by chi square statistics. We represent each feature by TF(text frequency), TF-IDF, and binary weight for experiments. After training SVM with the selected features, SVM classifies each e-mail as spam or not. In experiment, the selected features improve the performance of our system and we acquired overall 98.9% of accuracy with TREC05-p1 spam corpus.

• Korean Journal of Oriental Medicine
• /
• v.12 no.3 s.18
• /
• pp.1-15
• /
• 2006

For the recent several years, there has been much interest in lexical semantic network. However, it seems to be very difficult to evaluate the effectiveness and correctness of it and invent the methods for applying it into various problem domains. In order to offer the fundamental ideas about how to evaluate and utilize lexical semantic networks, we developed two automatic word clustering systems, which are called system A and system B respectively. 68,455,856 words were used to learn both systems. We compared the clustering results of system A to those of system B which is extended by the lexical-semantic network. The system B is extended by reconstructing the feature vectors which are used the elements of the lexical-semantic network of 3,656 '-ha' verbs. The target data is the 'multilingual Word Net-CoreNet'.When we compared the accuracy of the system A and system B, we found that system B showed the accuracy of 46.6% which is better than that of system A, 45.3%.

• The Journal of Information Systems
• /
• v.29 no.3
• /
• pp.237-251
• /
• 2020

Purpose The research was studied the hierarchical Hangul emotion index by organizing all the emotions which SNS users are thinking. As a preliminary study by the researcher, the English-based Plutchick (1980)'s emotional standard was reinterpreted in Korean, and a hashtag with implicit meaning on SNS was studied. To build a multidimensional emotion dictionary and classify three-dimensional emotions, an emotion seed was selected for the composition of seven emotion sets, and an emotion word dictionary was constructed by collecting SNS hashtags derived from each emotion seed. We also want to explore the priority of each Hangul emotion index. Design/methodology/approach In the process of transforming the matrix through the vector process of words constituting the sentence, weights were extracted using TF-IDF (Term Frequency Inverse Document Frequency), and the dimension reduction technique of the matrix in the emotion set was NMF (Nonnegative Matrix Factorization) algorithm. The emotional dimension was solved by using the characteristic value of the emotional word. The cosine distance algorithm was used to measure the distance between vectors by measuring the similarity of emotion words in the emotion set. Findings Customer needs analysis is a force to read changes in emotions, and Korean emotion word research is the customer's needs. In addition, the ranking of the emotion words within the emotion set will be a special criterion for reading the depth of the emotion. The sentiment index study of this research believes that by providing companies with effective information for emotional marketing, new business opportunities will be expanded and valued. In addition, if the emotion dictionary is eventually connected to the emotional DNA of the product, it will be possible to define the "emotional DNA", which is a set of emotions that the product should have.

• The Journal of the Acoustical Society of Korea
• /
• v.13 no.2E
• /
• pp.41-50
• /
• 1994

This paper proposes a DMS(Dynamic Multi-Section) model based on the information of the similar features in word pattern. This model represents each word as a time series of several sections and each section implies duration time information and typical feature vectors. The procedure to make a model in the word pattern is that typical feature vector and duration time information are reflected in the distance, when matching between word pattern and model is repeated. As the result of it, the accumulated distance by matching is to be minimized.