• Journal of Digital Convergence
• /
• v.14 no.8
• /
• pp.233-243
• /
• 2016

Due to speech recognition problems in noisy environment, Audio Visual Speech Recognition (AVSR) system, which combines speech information and visual information, has been proposed since the mid-1990s,. and lip reading have played significant role in the AVSR System. This study aims to enhance recognition rate of utterance word using only lip shape detection for efficient AVSR system. After preprocessing for lip region detection, Convolution Neural Network (CNN) techniques are applied for utterance period detection and lip shape feature vector extraction, and Hidden Markov Models (HMMs) are then used for the recognition. As a result, the utterance period detection results show 91% of success rates, which are higher performance than general threshold methods. In the lip reading recognition, while user-dependent experiment records 88.5%, user-independent experiment shows 80.2% of recognition rates, which are improved results compared to the previous studies.

• Journal of the Institute of Convergence Signal Processing
• /
• v.10 no.3
• /
• pp.170-175
• /
• 2009

In this paper, we propose real time automatic speaker verification (ASV) system to protect personal information on personal digital assistant (PDA) device. Recently, the capacity of PDA has extended and been popular, especially for mobile environment such as mobile commerce (M-commerce). However, there still exist lots of difficulties for practical application of ASV utility to PDA device because it requires too much computational complexity. To solve this problem, we apply the method to relieve the computational burden by performing the preprocessing such as spectral subtraction and speech detection during the speech utterance. Also by applying the hidden Markov model (HMM) optimal state alignment and the sequential probability ratio test (SPRT), we can get much faster processing results. The whole system implementation is simple and compact enough to fit well with PDA device's limited memory and low CPU speed.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.36S no.9
• /
• pp.103-112
• /
• 1999

In this paper, a speaker-independent isolated world recognition system for a car navigation system is implemented using a general digital signal processor. This paper presents a method combining SNR normalization with RAS as a noise processing method. The semi-continuous hidden markov model is adopted and TMS320C31 is used in implementing the real-time system. Recognition word set is composed of 69 command words for a car navigation system. Experimental results showed that the recognition performance has a maximum of 93.62% in case of a combination of SNR normalization and spectral subtraction, and the performance improvement rate of the system is 3.69%, Presented noise processing method showed good speech recognition performance in 5dB SNR in car environment.

• Journal of Digital Convergence
• /
• v.11 no.11
• /
• pp.367-372
• /
• 2013

In general, Levenshtein distance algorithm have a problem with not distinguish the consideration of vacabulary retrieve, because Levenshtein methode is used to vocabulary order are not defined. In this paper, we propose a improved Levenshtein methode, it effectively manage the vocabulary retrieve by frequency use of a vocabulary, and it gives the weight number which have a order between vocabularies. Therefore proposed methode have a advantage of solve the defect of perception rate in the case of increase the vocabulary, improve the recognition time become higher and it can be effectively retrieval space management.. System performance as a result of represent vocabulary dependence recognition rate of 97.81%, vocabulary independence recognition rate of 96.91% in indoor environment. Also, vocabulary dependence recognition rate of 91.11%, vocabulary independence recognition rate of 90.01% in outdoor environment.

• The Journal of Bigdata
• /
• v.1 no.1
• /
• pp.61-66
• /
• 2016

With big data analysis, companies use the customized marketing strategy based on customer's information. However, because of the concerns about privacy issue and identity theft, people start erasing their personal information or changing the privacy settings on social network site. Facebook, the most used social networking site, has the feature called 'Likes' which can be used as a tool to predict user's demographic profiles, such as sex and age range. To make accurate analysis model for the study, 'Likes' data has been processed by using Gaussian RBF and nFactors for dimensionality reduction. With random Forest and 5-fold cross-validation, the result shows that sex has 75% and age has 97.85% accuracy rate. From this study, we expect to provide an useful guideline for companies and marketers who are suffering to collect customers' data.

• Journal of IKEEE
• /
• v.20 no.1
• /
• pp.9-15
• /
• 2016

The endpoint detection (EPD) is the most important technique in plasma etching process. In plasma etching process, the Optical Emission Spectroscopy (OES) is usually used to analyze plasma reaction. And Plasma Impedance Monitoring (PIM) system is used to measure the voltage, current, power, and load impedance of the supplied RF power during plasma process. In this paper, a new decision making algorithm is proposed to improve the performance of EPD in SiOx single layer plasma etching. To enhance the accuracy of the endpoint detection, both OES data and PIM data are utilized and a newly proposed decision making algorithm is applied. The proposed method successfully detected endpoint of silicon oxide plasma etching.

• Proceedings of the Acoustical Society of Korea Conference
• /
• autumn
• /
• pp.29-32
• /
• 2000

본 논문에서는 음성인식 시스템의 음향모델 개선을 위한 기초적 연구로서, 문맥적인 요소를 필요로 하는 SSS(Successive State Splitting)와 필요로 하지 않는 SSS-free 알고리즘을 이용한 HMnet(Hidden Markov Network) 음향모델 작성방법에 대해 검토하고 작성한 음향모델을 한국어에 적용하여 그 유효성을 확인하였다. HMnet을 이용한 음소모델의 작성방법은 전체 학습 데이터에 대해서 각각 2개의 상태를 가지는 초기 모델을 작성한 후, 이를 시간과 문맥방향으로의 최대 분포를 가지는 상태를 재분할한 후 임의의 상태수가 될 때까지 상태분할을 계속적으로 수행케 하여 각 음소모델을 작성하게 된다. 작성한 HMnet 음향모델의 유효성을 확인하기 위해 ETRI 445 단어의 3인에 대한 화자종속 음소인식 실험을 수행하였다. 인식실험 결과, SSS 알고리즘을 이용한 화자종속실험의 경우 상태수 520에서 평균 $62.8\%$의 인식률을, SSS-free 알고리즘의 경우 상태수 420에서 평균 $64.2\%$의 인식률을 얻었다. 이 결과는 HMM을 이용한 경우(약$43.4\%$)보다 $20\%$이상의 인식률 향상을 보여 이 알고리즘의 유효성을 확인할 수 있었다. SSS와 SSS-free를 비교한 경우, SSS-free가 SSS보다 낮은 상태수에서 평균 $1.4\% 향상된 인식률을 보였다.

• 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.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.9 no.11
• /
• pp.1221-1226
• /
• 2014

In an effort to improve the efficiency of the speech recognition, we investigate the effect of the number of phoneme clusters. For this purpose, codebooks of varied number of phoneme clusters are prepared by modified k-means clustering algorithm. The subsequent processing is fuzzy vector quantization (FVQ) and hidden Markov model (HMM) for speech recognition test. The result shows that there are two distinct regimes. For large number of phoneme clusters, the recognition performance is roughly independent of it. For small number of phoneme clusters, however, the recognition error rate increases nonlinearly as it is decreased. From numerical calculation, it is found that this nonlinear regime might be modeled by a power law function. The result also shows that about 166 phoneme clusters would be the optimal number for recognition of 300 isolated words. This amounts to roughly 3 variations per phoneme.

• Proceedings of the Korea Institute of Convergence Signal Processing
• /
• 2003.06a
• /
• pp.6-9
• /
• 2003

In this paper, we studied on the difference of speaker adaptation according to the phoneme classification for Korean Speech recognition. In order to study of speech adaptation according to the weight of difference of phoneme as recognition unit, we used SCHMM as recognition system. And Speaker adaptation method used in this paper was MAPE(Maximum A Posteriori Probability Estimation), Linear Spectral Estimation. In order to evaluate the performance of these methods, we used 10 Korean isolated numbers as the experimental data. It is possible for the first and the second methods to be carried out unsupervised learning and used in on-line system. And the first method was shown performance improvement over the second method, and hybrid adaptation showed the better recognition results than those which performed each method. And the result of Speaker adaptation using the variable weight according to the phoneme had better than the result using fixed weight.