• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.35 no.7C
• /
• pp.604-610
• /
• 2010

In this paper we proposed FSFN (Filter bank sub-band energy subtraction based CLSFN) method to improve the recognition performance of the existing CLSFN (Cepstral distance and Log-energy based Silence Feature Normalization). The proposed FSFN reduces the energy of noise components in filter bank sub-band domain when extracting the features from speech data. This leads to extract the enhanced cepstral features and thus improves the accuracy of speech/silence classification using the enhanced cepstral features. Therefore, it can be expected to get improved performance comparing with the existing CLSFN. Experimental results conducted on Aurora 2.0 DB showed that our proposed FSFN method improves the averaged word accuracy of 2% comparing with the conventional CLSFN method, and FSFN combined with CMVN (Cepstral Mean and Variance Normalization) also showed the best recognition performance comparing with others.

• Korean Journal of Ichthyology
• /
• v.8 no.2
• /
• pp.57-67
• /
• 1996

To investigate the fishes of the Arafura Sea of Indonesia, fish specimens were collected by scientists on board Korean trawlers fishing in this region during June 1994 and March 1996. A total of 137 fish species were identified, of which 40 occurred in both 1994 and 1996. In cartilaginous fishes, fishes from the families Carcharhinidae and Dasyatididae were dominant in terms of number of species and in teleost fishes, dominant species were those from the families Engraulididae, Carangidae, Sciaenidae and Tetraodontidae. This study revealed that Sardinelia longiceps, Setipinna melanochir, Cypselurus hiraii, Podothecus sachi, Nemipterus aurora, Johnius grypotus, Moolgarda perusi were collected for the first time in the Arafura Sea.

• Korea Information Processing Society Review
• /
• v.15 no.2
• /
• pp.15-28
• /
• 2008

The Korea UK e-Science Collaboration project has been aimed at supporting research and collaboration between Korean and UK researchers. Its goal is to benefit the nation by reducing cost and time in constructing the National e-Science Research Environment' by studying and benchmarking the cases in countries that have already procured advanced technologies in the area. Two joint workshops were held in a year where researchers from the two countries had the opportunities to share their research results with each other. Also, the project has supported exchanges of researchers fostering expertise in the field. In the course of the project, the e-Science Centre in the UK and KISTI have signed MoU(Memorandum of Understanding) in 2006. Moreover, there have been active research collaboration between Korea and the UK. The University of Southampton will share the BioSimGrid data with the Korean counterpart, and the University of York has provided the AURA software. In the future, KISTI and the UK NGS(National Grid Service) will organize a working group at OGF that will work mainly on the standardization of Parameter Sweep and bring it to lead the global standard. KISTI will include its own AURORA system into OMI-UK software stack, which will enable access to NGS resources through AURORA user application. The collaboration with the UK has opened up more opportunities for collaboration with other countries as well. KISTI and HLRS in Germany have agreed to share the COVISE and will have research exchanges. As such, it is expected that Korea will play a major role in e-Science research by building strategic and systematic collaborative relations with its International partners.

• The Journal of the Acoustical Society of Korea
• /
• v.34 no.4
• /
• pp.316-320
• /
• 2015

In this paper, the pole filtering concept is applied to the Mel-frequency cepstral coefficient (MFCC) feature vectors in the conventional cepstral mean normalization (CMN) and cepstral mean and variance normalization (CMVN) frameworks. Additionally, performance of the cepstral mean and scale normalization (CMSN), which uses scale normalization instead of variance normalization, is evaluated in speech recognition experiments in noisy environments. Because CMN and CMVN are usually performed on a per-utterance basis, in case of short utterance, they have a problem that reliable estimation of the mean and variance is not guaranteed. However, by applying the pole filtering and scale normalization techniques to the feature normalization process, this problem can be relieved. Experimental results using Aurora 2 database (DB) show that feature normalization method combining the pole-filtering and scale normalization yields the best improvements.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.9 no.5
• /
• pp.543-548
• /
• 2014

Multiple-model based speech recognition framework (MMSR) has been known to be very successful in speech recognition. Since it uses multiple hidden Markov modes (HMMs) that corresponds to various noise types and signal-to-noise ratio (SNR) values, the selected acoustic model can have a close match with the test noisy speech. However, since the number of HMM sets is limited in practical use, the acoustic mismatch still remains as a problem. In this study, we experimentally determined the optimal SNR mapping between the test noisy speech and the HMM set to mitigate the mismatch between them. Improved performance was obtained by employing the SNR mapping instead of using the estimated SNR from the test noisy speech. When we applied the proposed method to the MMSR, the experimental results on the Aurora 2 database show that the relative word error rate reduction of 6.3% and 9.4% was achieved compared to a conventional MMSR and multi-condition training (MTR), respectively.

• The Journal of the Acoustical Society of Korea
• /
• v.37 no.6
• /
• pp.506-511
• /
• 2018

This paper proposes an GMM(Gaussian Mixture Model)-DNN(Deep Neural Network) hybrid-based feature compensation method for effective speech recognition in noisy environments. In the proposed algorithm, the posterior probability for the conventional GMM-based feature compensation method is calculated using DNN. The experimental results using the Aurora 2.0 framework and database demonstrate that the proposed GMM-DNN hybrid-based feature compensation method shows more effective in Known and Unknown noisy environments compared to the GMM-based method. In particular, the experiments of the Unknown environments show 9.13 % of relative improvement in the average of WER (Word Error Rate) and considerable improvements in lower SNR (Signal to Noise Ratio) conditions such as 0 and 5 dB SNR.

• ETRI Journal
• /
• v.36 no.5
• /
• pp.835-840
• /
• 2014

New ways of achieving small, multiband, multifunctional, and standard solutions for mobile handset antennas are demanded in the current wireless market. A non-resonant element of $5mm{\times}5mm{\times}5mm$, a matching network, and a $100mm{\times}40mm$ slotted ground plane are proposed to satisfy mobile market demands that require multiband operation and small antenna solutions. The main advantage of the proposed design is that with only one non-resonant element of considerably small size ($0.015{\lambda}$, 900 MHz), the handset is capable of providing operation at mobile bands.

• MALSORI
• /
• no.62
• /
• pp.85-96
• /
• 2007

In this paper, we carried out recognition experiments for noisy speech having various levels of car noise and output of an audio system using the speech interface. The speech interface consists of three parts: pre-processing, acoustic echo canceller, post-processing. First, a high pass filter is employed as a pre-processing part to remove some engine noises. Then, an echo canceller implemented by using an FIR-type filter with an NLMS adaptive algorithm is used to remove the music or speech coming from the audio system in a car. As a last part, the MMSE-STSA based speech enhancement method is applied to the out of the echo canceller to remove the residual noise further. For recognition experiments, we generated test signals by adding music to the car noisy speech from Aurora 2 database. The HTK-based continuous HMM system is constructed for a recognition system. Experimental results show that the proposed speech interface is very promising for robust speech recognition in a noisy car environment.

• Speech Sciences
• /
• v.11 no.1
• /
• pp.7-20
• /
• 2004

In this paper, we propose a two-step noise compensation algorithm in feature extraction for achieving robust speech recognition. The proposed method frees us from requiring a priori information on noisy environments and is simple to implement. First, in frequency domain, the Harmonics-based Spectral Subtraction (HSS) is applied so that it reduces the additive background noise and makes the shape of harmonics in speech spectrum more pronounced. We then apply a judiciously weighted variance Feature Vector Normalization (FVN) to compensate for both the channel distortion and additive noise. The weighted variance FVN compensates for the variance mismatch in both the speech and the non-speech regions respectively. Representative performance evaluation using Aurora 2 database shows that the proposed method yields 27.18% relative improvement in accuracy under a multi-noise training task and 57.94% relative improvement under a clean training task.

• Proceedings of the Acoustical Society of Korea Conference
• /
• autumn
• /
• pp.139-142
• /
• 2004

훈련 과정과 인식 과정사이의 주변 잡음과 채널 특성으로 인한 환경의 불일치는 음성 인식 성능을 급격히 저하시킨다. 이러한 차이를 극복하기 위해 다양한 전처리 방법이 제안되어 왔으며, 최근에는 스테레오 데이터와 잡음 음성의 Gaussian Mixture Model(GMM)을 이용하여 보상벡터를 구하는 SPLICE 방법이 좋은 성능을 보여주고 있다. 하지만 차원별로 특징벡터를 보상해주는 추정된 보상벡터는 underestimation되는 경향이 있으며, 그 정도가 각각의 차원마다 달라짐이 관찰되었다. 본 논문에서는 SPLICE 방법에 기반하여 추정된 보상벡터와 실제 보상벡터 사이의 관계를 관찰하여 차원별로 다른 가중치를 적용하는 차원별 가중 보상 방법을 제안하였다. 제안한 방법은 Aurora2 Clean-condition인 경우 baseline 실험 결과에 비해 $68\%$의 높은 상대적인 인식 향상율을 얻었다.