• The Journal of the Acoustical Society of Korea
• /
• v.10 no.1
• /
• pp.52-60
• /
• 1991

This paper describes the study on isolated word recognition by using DHMM(Dynamic Hidden Markov Model) which has dynamic feature of spectrum as a parameter. This paper discusses speech recognition experiment basedon HMM which can evaluate not only instantaneous spectral features but also dynamic spectral features. LPC cepstrum parameters is used as a static feature and LPC cepstrum's regression coefficient is used as a dynamic feature. These two features are quantized by each VQ codebook. DHMM is modeled by receiving static vector and dynamic vector by input. In the whole experiment, as recognition experiment using DHMM shows 92.7% of recognition rate while the experiment using conventional HMM shows 88.8% of recognition rate, DHMM proved to be a useful model.

• The Journal of the Acoustical Society of Korea
• /
• v.16 no.2
• /
• pp.71-79
• /
• 1997

In this paper, a codeword-dependent distance normalization(CDDN) and an instar-formed fuzzy smoothing of output distribution are proposed for robust estimation of output probabilities in the FVQ(fuzzy vector quantization)-DHMM(discrete hidden Markov model). The FVQ-DHMM is a variant of DHMM in which the state output probability is estimated by the sum oft he product of the output probability and its weighting factor for each codeword on an input vector. As the performance of the FVQ-DHMM is influenced by weighting factor and output distribution from a state, it is required to get a method to get robust estimation of weighting factors and output distribution for each state. From experimental results, the proposed CDDN method has reduced 24% of error rate over the conventional FVQ-DHMM, and also reduced 79% of error rate when the smoothing of output distribution is also applied to the computation of an output probability. These results indicate that the use of CDDN and the fuzzy smoothing of output distribution to the FVQ-DHMM lead to improved recognition, and therefore it may be used as an alternative to the robust estimation of output probabilities for HMMs.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2002.05a
• /
• pp.171-174
• /
• 2002

음성 신호는 언어정보, 개인성, 감정 등의 여러 가지 정보를 포함한 음향학적인 신호인 동시에 가장 자연스럽고 널리 쓰이는 의사소통 수단의 하나이다. 본 연구에서는 저장된 음성 신호에서 추출한 특징 파라미터를 사용한 경우와 음성 특징파라미터에 입술 패턴에 대한 영상정보를 통시에 사용한 경우 DHMM과 신경망을 통하여 각각 인식률을 비교해 보았다. 그 결과 입술패턴에 대할 영상정보도 음성인식에 사용 할 수 있음을 알 수 있었다.

• The Journal of the Acoustical Society of Korea
• /
• v.19 no.4
• /
• pp.23-30
• /
• 2000

In this paper, the web-based speaker verification system is designed. To decide the speaker recognition algorithm applied to the web-based speaker verification system, the recognition performance and special features of the text-dependent speaker recognition algorithms(DTW, DHMM, SCHMM) are compared through the computer simulation. Using the results of computer simulation, select DHMM as speaker recognition algorithm at web-based speaker verification system because DHMM has the proper recognition performance and initial training utterance number. And by the three-tier method using the ActiveX, DCOM techniques web-based speaker verification system is designed to be operated in the distributed processing environment.

• Proceedings of the Korean Information Science Society Conference
• /
• 2000.10b
• /
• pp.374-376
• /
• 2000

음성인식 시스템에 사용하는 알고리즘 중에 하나인 DHMM은 코드북을 이용하여 음성의 프레임들에 대한 특징을 관측열로 추출하여 음성의 패턴에 대한 훈련과 인식을 수행하게 된다. 그러나 음성은 유성음과 무성음의 특징 차이가 많이 나게 되므로 하나의 코드북을 이용하게 되면 코드북 오차에 의하여 성질이 전혀 다른 코드북 인덱스를 DHMM의 관측열로 사용하게 된다. 본 논문에서는 음성의 유성음과 무성음에 대한 선별적인 작업을 통해 서로 다른 코드북을 만들어 관측열을 추출하고 선행 관측과 현 관측과의 거리 비교 연산을 통하여 관측의 시간축을 정규화한 관측열을 음성인식에 사용하였다. 본 논문에서 제시하는 인식 방법을 사용하여 실험한 결과, 기존의 인식 방법보다 5.33% 향상된 결과를 얻었다.

• 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%.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2009.10a
• /
• pp.99-105
• /
• 2009

Crack detection method of a rotating blade was suggested in this paper. A rotating blade was modeled with a cantilever beam connected to a hub undergoing rotating motion. The existence and the location of crack were able to be recognized from the vertical response of end tip of a rotating cantilever beam by employing Discrete Hidden Markov Model (DHMM) and Empirical Mode Decomposition (EMD). DHMM is a famous stochastic method in the field of speech recognition. However, in recent researches, it has been proved that DHMM can also be used in machine health monitoring. EMD is the method suggested by Huang et al. that decompose a random signal into several mono component signals. EMD was used in this paper as the process of extraction of feature vectors which is the important process to developing DHMM. It was found that developed DHMMs for crack detection of a rotating blade have shown good crack detection ability.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.28B no.7
• /
• pp.548-556
• /
• 1991

In this paper, Korean spoken continuous digits are ercognized using DHMM(Dynamic Hidden Markov Model) and lexical analysis to provide the base of developing voice dialing system. After segmentation by phoneme unit, it is recognized. This system can be divided into the segmentation section, the design of standard speech section, the recognition section, and the lexical analysis section. In the segmentation section, it is segmented using the ZCR, O order LPC cepstrum, and Ai, parameter of voice speech dectaction, which is changed according to time. In the standard speech design section, 19 phonemes or syllables are trained by DHMM and designed as a standard speech. In the recognition section, phomeme stream are recognized by the Viterbi algorithm.In the lexical decoder section, finally recognized continuous digits are outputed. This experiment shiwed the recognition rate of 85.1% using data spoken 7 times of 21 classes of 7 continuous digits which are combinated all of the occurence, spoken by 10 man.

• The Journal of the Acoustical Society of Korea
• /
• v.13 no.4
• /
• pp.40-49
• /
• 1994

In this paper, we propose methods for improving the performance of word recognition system. The ray stratey of the first method is to apply the inertia to the feature vector sequences of speech signal to stabilize the transitions between VQ cdoes. The second method is generating the new observation probabilities using the transition matrix of VQ codes as weights at the observation probability of the output symbol, so as to take into account the time relation between neighboring frames in DHMM. By applying the inertia to the feature vector sequences, we can reduce the overlapping of probability distribution of the response paths for each word and stabilize state transitions in the HMM. By using the transition matrix of VQ codes as weights in conventional DHMM. we can divide the probability distribution of feature vectors more and more, and restrict the feature distribution to a suitable region so that the performance of recognition system can improve. To evaluate the performance of the proposed methods, we carried out experiments for 50 DDD area names. As a result, the proposed methods improved the recognition rate by $4.2\%$ in the speaker-dependent test and $12.45\%$ in the speaker-independent test, respectively, compared with the conventional DHMM.

• The KSFM Journal of Fluid Machinery
• /
• v.7 no.2 s.23
• /
• pp.41-49
• /
• 2004

Condition monitoring is very important in turbo machinery because single failure could cause critical damages to its plant. So, automatic fault recognition has been one of the main research topics in condition monitoring area. We have used a relatively new fault recognition method, Hidden Markov Model(HMM), for mechanical system. It has been widely used in speech recognition, however, its application to fault recognition of mechanical signal has been very limited despite its good potential. In this paper, discrete HMM(DHMM) was used to recognize the faults of rotor system to study its fault recognition ability. We set up a rotor kit under unbalance and oil whirl conditions and sampled vibration signals of two failure conditions. DHMMS of each failure condition were trained using sampled signals. Next, we changed the setup and the rotating speed of the rotor kit. We sampled vibration signals and each DHMM was applied to these sampled data. It was found that DHMMs trained by data of one rotating speed have shown good fault recognition ability in spite of lack of training data, but DHMMs trained by data of four different rotating speeds have shown better robustness.