• Annual Conference on Human and Language Technology
• /
• 1999.10e
• /
• pp.167-172
• /
• 1999

구어는 프로그래밍 언어와는 달리 주어진 문장 내에서의 해당 어휘의 뜻(semantic information)을 알고 다른 어휘들과의 연관성 (grammatical information)을 알아야만 적절한 형태소분석이 가능하다. 또한 구어는 방대한 양의 어휘들로 구성되어 있으며 사용하는 사람마다의 다양한 응용과 공식화되기 어려운 수많은 예외들로 운용되기 때문에 단순히 찾아보기표와 오토마타만으로는 형태소분석에 한계가 있다. 이에 본 논문에서는 주어진 어휘집과 그 어휘들로 만들어진 다양한 문장들로부터 구어운용의 근본기제를 스스로 학습해나가는 강화학습중심의 언어모델을 제안하고 실제로 한국어 형태소분석에 적용하여 그 성능과 특성을 파악해보았다. 구어파서의 입력은 음절단위의 발음이며 인간이 문장을 듣거나 보는 것과 동일하게 시간에 따라 순차적으로 입력된다. 파서의 출력 또한 시간에 따라 변화되면서 나타나며 입력된 연속음절을 형태소단위로 분리(segmentation)하고 분류(labeling)한 결과를 나타낸다. 생성인식 언어모델이 기존의 언어모델과 다른 점은 구어 파싱에 있어서 필수적인 미등륵어에 대한 유연성과 앞단의 음성인식기 오류에 적절한 반응(fault tolerance)을 나타내는 것이다.

• The Journal of the Acoustical Society of Korea
• /
• v.19 no.3
• /
• pp.35-44
• /
• 2000

In this paper, we propose a new trajectory model for characterizing segmental features and their interaction based upon a general framework of hidden Markov models. Each segment, a sequence of vectors, is represented by a trajectory of observed sequences. This trajectory is obtained by applying a new design matrix which includes transitional information on contiguous frames, and is characterized as a polynomial regression function. To apply the trajectory to the segmental HMM, the frame features are replaced with the trajectory of a given segment. We also propose the likelihood of a given segment and the estimation of trajectory parameters. The obervation probability of a given segment is represented as the relation between the segment likelihood and the estimation error of the trajectories. The estimation error of a trajectory is considered as the weight of the likelihood of a given segment in a state. This weight represents the probability of how well the corresponding trajectory characterize the segment. The proposed model can be regarded as a generalization of a conventional HMM and a parametric trajectory model. The experimental results are reported on the TIMIT corpus and performance is show to improve significantly over that of the conventional HMM.

• The Journal of the Acoustical Society of Korea
• /
• v.21 no.4
• /
• pp.401-406
• /
• 2002

Each Korean digit is composed of only a syllable, so recognizers as well as Korean often have difficulty in recognizing it. When digit strings are pronounced, the original pronunciation of each digit is largely changed due to the co-articulation effect. In addition to these problems, the distortion caused by various channels and noises degrades the recognition performance of Korean connected digit string. This paper dealt with some techniques to improve recognition performance of it, which include defining a set of PLUs by considering phonemic variations in Korean digit and constructing a recognizer to handle speakers various speaking styles. In the speaker-independent connected digit recognition experiments using telephone speech, the proposed techniques with 1-Gaussian/state gave string accuracy of 83.2%, i. e., 7.2% error rate reduction relative to baseline system. With 11-Gaussians/state, we achieved the highest string accuracy of 91.8%, i. e., 4.7% error rate reduction.

• The Journal of the Korea Contents Association
• /
• v.5 no.2
• /
• pp.147-155
• /
• 2005

In this paper, we present a FAES(a Facial Animation with Emotion and Speech) system for speech-driven face animation with emotions. We animate face cartoons not only from input speech, but also based on emotions derived from speech signal. And also our system can ensure smooth transitions and exact representation in animation. To do this, after collecting the training data, we have made the database using SVM(Support Vector Machine) to recognize four different categories of emotions: neutral, dislike, fear and surprise. So that, we can make the system for speech-driven animation with emotions. Also, we trained on Korean young person and focused on only Korean emotional face expressions. Experimental results of our system demonstrate that more emotional areas expanded and the accuracies of the emotional recognition and the continuous speech recognition are respectively increased 7% and 5% more compared with the previous method.

• Journal of KIISE
• /
• v.45 no.1
• /
• pp.22-29
• /
• 2018

Over the decades, neural networks have been successfully used in numerous applications from speech recognition to image classification. However, these neural networks cannot explain their results and one needs to know how and why a specific conclusion was drawn. Most studies focus on extracting binary rules from neural networks, which is often impractical to do, since data sets used for machine learning applications contain continuous values. To fill the gap, this paper presents an algorithm to extract logic rules from a trained neural network for data with continuous attributes. It uses hyperplane-based linear classifiers to extract rules with numeric values from trained weights between input and hidden layers and then combines these classifiers with binary rules learned from hidden and output layers to form non-linear classification rules. Experiments with different datasets show that the proposed approach can accurately extract logical rules for data with nonlinear continuous attributes.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1998.06c
• /
• pp.71-74
• /
• 1998

HMM (hidden Markov model)을 이용한 음성인식은 현재 가장 널리 쓰여지고 있는 방법으로, 이 중 CDHMM (continuous observation density HMM)은 상태에서 관측심볼확률을 연속확률밀도를 사용하여 표현한다. 본 논문에서는 가우스 혼합밀도함수를 사용하는 CDHMM의 상태천이확률과, 관측심볼확률을 표현하기 위한 인자인 평균벡터, 공분산 행렬, 가지하중값을 유전자 알고리듬을 사용하여 최적화하는 방법을 제안하였다. 유전자 알고리듬은 매개변수 최적화문제에 대하여 자연의 진화원리를 모방한 알고리듬으로, 염색체 형태로 표현된 개체군 (population) 중에서 환경에 대한 적합도 (fitness)가 높은 개체가 높은 확률로 살아남아 재생 (reproduction)하게 되며, 교배 (crossover)와 돌연변이 (mutation) 연산 후에 다음 세대 개체군을 형성하게 되고, 이러한 과정을 반복하면서 최적의 개체를 구하게 된다. 본 논문에서는 상태천이확률, 평균벡터, 공분산행렬, 가지하중값을 부동소수점수 (floating point number)의 유전자형으로 표현하여 유전자 알고리듬을 수행하였다. 유전자 알고리듬은 복잡한 탐색공간에서 최적의 해를 찾는데 효과적으로 적용되었다.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.17 no.5
• /
• pp.472-481
• /
• 1992

In this paper, we proposed a method of segmenting speech signal by neural network and its validity is proved by computer simulation. The neural network Is composed of multi layer perceptrons with one hidden layer. The matching accuracies of the proposed algorithm are measured for continuous vowel and place names. The resulting average matching accuracy is 100% for speaker-dependent case, 99.5% for speaker-independent case and 94.5% for each place name when the neural network 1,; trained for 6 place names simultaneously.

• 제어로봇시스템학회:학술대회논문집
• /
• 1993.10a
• /
• pp.870-876
• /
• 1993

This paper presents a binary clustering network (BCN) and a heuristic algorithm to detect pitch for recognition of keywords in continuous speech. In order to classify nonlinear patterns, BCN separates patterns into binary clusters hierarchically and links same patterns at root level by using the supervised learning and the unsupervised learning. BCN has many desirable properties such as flexibility of dynamic structure, high classification accuracy, short learning time, and short recall time. Pitch Detection algorithm is a heuristic model that can solve the difficulties such as scaling invariance, time warping, time-shift invariance, and redundance. This recognition algorithm has shown recognition rates as high as 95% for speaker-dependent as well as multispeaker-dependent tests.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.32B no.12
• /
• pp.1635-1643
• /
• 1995

This paper proposes an algorithm for machine recognition of phonemes in continuously spoken Korean. The proposed algorithm is a static strategy neural network. The algorithm uses, at the stage of training neurons, features such as the rate of zero crossing, short-term energy, and either PARCOR or auditory-like perceptual linear prediction(PLP) but not both, covering a time of 171ms long. Numerical results show that the algorithm with PLP achieves approximately the frame-based phoneme recognition rate of 99% for small vocabulary recognition experiments. Based on this it is concluded that the proposed algorithm with PLP analysis is effective in phoneme recognition.

• Proceedings of the KIEE Conference
• /
• 1988.07a
• /
• pp.624-627
• /
• 1988

This paper describes the implementation of recognition of speaker - dependent Korean spoken continuous digits. The recognition system can be divided into two parts, acoustic - phonetic processor and lexical decoder. Acoustic - phonetic processor calculates the feature vectors from input speech signal and the performs frame labelling and phone labelling. Frame labelling is performed by Bayesian classification method and phone labelling is performed using labelled frame and posteriori probability. The lexical decoder accepts segments (phones) from acoustic - phonetic processor and decodes its lexical structure through phone network which is constructed from phonetic representation of ten digits. The experiment carried out with two sets of 4continuous digits, each set is composed of 35 patterns. An evaluation of the system yielded a pattern accuracy of about 80 percent resulting from a word accuracy of about 95 percent.