• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.22 no.5
• /
• pp.1026-1035
• /
• 1997

In this paper, we studied for phoneme recognition using GPFN and PNN as a kind of RBFN. The structure of RBFN is similar to a feedforward networks but different from choosing of activation function, reference vector and learnign algorithm in a hidden layer. Expecially sigmoid function in PNN is replaced by one category included exponential function. And total calculation performance is high, because PNN performs pattern classification with out learning. In phonemerecognition experiment with 5 vowel and 12 consant, recognition rates of GPFN and PNN as a kind of RBFN reflected statistic characteristic of speech are higher than ones of MLP in case of using test data and quantizied data by VQ and LVQ.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.24 no.2
• /
• pp.117-122
• /
• 2014

The minimal pairs mean the pairs that have same phonotactics except just one sound in the sequences cause different lexical items. This paper proposes the searching tool of minimal pairs for efficiency of phonological researches with minimal pairs. We suggest a guide to develop Korean minimal pair searching programs by comparing to other programs. Proposing tool has user-friendly interface, minimizing key inputs, for linguistics who are not fluent in computer programs. And it serves the function which classifies the words in dictionary for the detailed researches. And for efficiency, it increases speed of dictionary loading by separating syllables through Unicode analysis, and optimizes dictionary structure for searching efficiency. The searching algorithm gains in speed by hashing algorithm using syllable counts. In our tool, the speed is improved more than earlier version about 5 times at converting dictionary and about 3 times at searching.

• The Journal of the Acoustical Society of Korea
• /
• v.25 no.7
• /
• pp.312-318
• /
• 2006

The conventional lip-synch system has a two-step process, speech segmentation and recognition. However, the difficulty of speech segmentation procedure and the inaccuracy of training data set due to the segmentation lead to a significant Performance degradation in the system. To cope with that, the connected vowel recognition method using Head-Body-Tail (HBT) model is proposed. The HBT model which is appropriate for handling relatively small sized vocabulary tasks reflects co-articulation effect efficiently. Moreover the 7 vowels are merged into 3 classes having similar lip shape while the system is optimized by employing a class dependent SCHMM structure. Additionally in both end sides of each word which has large variations, 8 components Gaussian mixture model is directly used to improve the ability of representation. Though the proposed method reveals similar performance with respect to the CHMM based on the HBT structure. the number of parameters is reduced by 33.92%. This reduction makes it a computationally efficient method enabling real time operation.

• The Journal of the Acoustical Society of Korea
• /
• v.12 no.4
• /
• pp.57-62
• /
• 1993

본 논문은 신경망을 이용한 우리말 모음에 대한 인식 연구이다. 음성을 나누거나. 음소별 인식이나, 시간 신축 방법을 사용하지 않고 모음을 인식하였다. 식나의 변화에 따른 음성의 변화를 정적인 음성으로 취급하였다. 10개로 균등히 나눈 프레임에 각 프레임마다 10차의 PARCOR계수를 추출하였다. 신경망의 구조를 간단히 하기 위해서 단모음과 복모음을 구분하여 학습시켰으며, 출력 노드의 수를 감소시키기 위해 이진 코드 형태로 구성하였다.

• Journal of KIISE:Software and Applications
• /
• v.27 no.3
• /
• pp.290-301
• /
• 2000

Most morphological analysis methods require repetitive procedures of input character code conversion, segmentation and lemmatization of constituent morphemes, filtering of candidate results through looking up lexicons, which causes run-time inefficiency. To alleviate such problem of run-time inefficiency, many systems have introduced the notion of 'pre-analysis' of words. However, this method based on pre-analysis dictionary of surface also has a critical drawback in its practical application because the size of the dictionaries increases indefinite to cover all words. This paper hybridizes both extreme approaches methodologically to overcome the problems of the two, and presents a method of morphological analysis based on pre-analysis of partial words. Under such hybridized scheme, most computational overheads, such as segmentation and lemmatization of morphemes, are shifted to building-up processes of the pre-analysis dictionaries and the run-time dictionary look-ups are greatly reduced, so as to enhance the run-time performance of the system. Moreover, additional computing overheads such as input character code conversion can also be avoided because this method relies upon no graphemic processing.

• Proceedings of the KSPS conference
• /
• 2006.05a
• /
• pp.37-40
• /
• 2006

In this paper, we propose a VQ codebook design of speech recognition feature parameters in order to improve the performance of a distributed speech recognition system. For the context-dependent HMMs, a VQ codebook should be correlated with phonetic distributions in the training data for HMMs. Thus, we focus on a selection method of training data based on phonetic distribution instead of using all the training data for an efficient VQ codebook design. From the speech recognition experiments using the Aurora 4 database, the distributed speech recognition system employing a VQ codebook designed by the proposed method reduced the word error rate (WER) by 10% when compared with that using a VQ codebook trained with the whole training data.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2015.07a
• /
• pp.155-158
• /
• 2015

단어의 유사도 측정 알고리즘은 DB 인덱싱, 필터링, 소스코드 분석 소프트웨어, 음성 인식 등 다양한 분야에서 활용되고 있다. 하지만 기존의 단어의 유사도만 비교하는 시스템에는 발음이 비슷한 유사단어나 오타가 있는 유사단어들은 측정을 못하는 단점이 있다. 언어의 유사도 측정에서는 알파벳만으로 볼게 아니라 언어 발음의 발화적 특성 또한 고려되어야 한다. 본 논문에서는 글로벌 시장에서의 다국적 기업들의 제품이나 문화 수출 등의 도움이 되는 각 나라의 금기어와의 발화적 특성까지 고려한 단어 유사도를 측정 할 수 있는 시스템을 제안한다. 11개국의 4개 언어 총 21487개의 금기어 단어를 금기어 데이터로 사용하였다. 제안하는 방법의 성능을 평가하기 위하여 타 알고리즘과의 성능비교와 여러 나라의 다양한 언어의 사용자들로부터 사용자 평가를 수행하였고 제안하는 방법이 발음 유사도를 측정하지 않는 알고리즘보다 우수한 성능을 보임을 확인하였다.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.35S no.8
• /
• pp.96-105
• /
• 1998

The aim of this paper is to propose the method of WDHMM(Weighted DHMM), using the MLP-VQ for the improvement of speaker-independent connect-digit recognition system. MLP neural-network output distribution shows a probability distribution that presents the degree of similarity between each pattern by the non-linear mapping among the input patterns and learning patterns. MLP-VQ is proposed in this paper. It generates codewords by using the output node index which can reach the highest level within MLP neural-network output distribution. Different from the old VQ, the true characteristics of this new MLP-VQ lie in that the degree of similarity between present input patterns and each learned class pattern could be reflected for the recognition model. WDHMM is also proposed. It can use the MLP neural-network output distribution as the way of weighing the symbol generation probability of DHMMs. This newly-suggested method could shorten the time of HMM parameter estimation and recognition. The reason is that it is not necessary to regard symbol generation probability as multi-dimensional normal distribution, as opposed to the old SCHMM. This could also improve the recognition ability by 14.7% higher than DHMM, owing to the increase of small caculation amount. Because it can reflect phone class relations to the recognition model. The result of my research shows that speaker-independent connected-digit recognition, using MLP-VQ and WDHMM, is 84.22%.

• Phonetics and Speech Sciences
• /
• v.12 no.4
• /
• pp.81-90
• /
• 2020

This paper describes a method of building Korean conversational speech data in the emergency medical domain and proposes an annotation method for the collected data in order to improve speech recognition performance. To suggest future research directions, baseline speech recognition experiments were conducted by using partial data that were collected and annotated. All voices were recorded at 16-bit resolution at 16 kHz sampling rate. A total of 166 conversations were collected, amounting to 8 hours and 35 minutes. Various information was manually transcribed such as orthography, pronunciation, dialect, noise, and medical information using Praat. Baseline speech recognition experiments were used to depict problems related to speech recognition in the emergency medical domain. The Korean conversational speech data presented in this paper are first-stage data in the emergency medical domain and are expected to be used as training data for developing conversational systems for emergency medical applications.