• Journal of Korea Multimedia Society
• /
• v.11 no.11
• /
• pp.1471-1479
• /
• 2008

In this paper, we propose a technique to improve Out-Of-Vocabulary(OOV) rejection algorithms in variable vocabulary recognition system which is much used in ASR(Automatic Speech Recognition). The rejection system can be classified into two categories by their implementation method, keyword spotting method and utterance verification method. The utterance verification method uses the likelihood ratio of each phoneme Viterbi score relative to anti-phoneme score for deciding OOV. In this paper, we add speaker verification system before utterance verification and calculate an speaker verification probability. The obtained speaker verification probability is applied for determining the proposed variable-confidence threshold. Using the proposed method, we achieve the significant performance improvement; CA(Correctly Accepted for keyword) 94.23%, CR(Correctly Rejected for out-of-vocabulary) 95.11% in office environment, and CA 91.14%, CR 92.74% in noisy environment.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.39C no.12
• /
• pp.1346-1351
• /
• 2014

We propose a novel approach for stereophonic acoustic echo suppression using spectro-temporal filtering based on soft decision. Unlike the conventional approaches estimating the echo pathes directly, the proposed technique can estimate stereo echo spectra without any double-talk detector. In order to improve the estimation of echo spectra, the extended power spectrum density matrix and echo overestimation control matrix are applied on this method. In addition, this echo suppression technique is based on soft decision technique using speech absence probability in STFT domain. Experimental results show that the proposed method improves compared with the conventional approaches.

• Parasites, Hosts and Diseases
• /
• v.34 no.2
• /
• pp.107-112
• /
• 1996

A nationwide survey was performed to know the geographical distribution of Gymnophalloides seoi (Digenea: Gymnophallidae) metacercariae in Korea, by examining the infection status of locally produced oysters, Crassosden gillu. A total of 24 coastal areas (myons) of 14 guns (=counties) in Kyonggi-do, Chollabuk-do, Chollanam-do, Kyongsangnam-do, Kyongsangbuk-do, or Kangwon-do, where natural oysters are produced but G. seoi has never been reported, and 13 areas (myons) of Shinan-gun, Chollanam-do, nearby the known endemic area, were surveyed. Oysters from non- endemic areas were free from G. seoi infection, except Byonsan-myon of Buan-gun, Chollabuk-do, where one of 50 oysters examined was infected with 15 metacercariae of G. seoi. In Shinan-gun, oysters from 10 areas including Aphae-myon (= township) and Anjwa-myon were infected with the metacercariae, with the infection rate ranging from 1.7% to 100% by areas. The intensity of infection was the highest in Aphae-myon, 785.9 metacercariae per oyster. The results indicate that high prevalence of G. seoi is confined to Shinan-gun, but low grade prevalence is also present in adjacent areas such as Buan- gun, Chollabuk-do.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.62 no.12
• /
• pp.1744-1749
• /
• 2013

Acoustic feedback is perceived as whistling or howling, which is a major complaint of hearing-aids users. Acoustic feedback cancellation is important in hearing-aids because acoustic feedback degrades performance of the hearing aid device by reducing maximum insertion gain. Adaptive systems for estimate acoustic feedback path and feedback suppression algorithms have been proposed in order to solve this problem. A typical feedback cancellation algorithm is LMS(least mean squares) because of its computational efficiency. However it has problem of convergence performance in high correlated input signal. In this paper, we propose a new variable step-size normalized LMS(least mean squares) algorithm using VAD(voice activity detection) to overcome the limitation of the LMS algorithm. The VAD algorithm is GSAP(global speech absence probability) and the feedback cancellation algorithm is normalized LMS. The proposed algorithm applies different step-size between voice and non-voice using VAD, for high stability, fast convergence speed and low misalignment when correlated inputs, such as speech. The result of simulation with white noise mixed speech signal, the proposed algorithm shows high performance then traditional algorithm in terms of stability, convergence speed and misalignment.

• The Journal of the Acoustical Society of Korea
• /
• v.10 no.1
• /
• pp.20-29
• /
• 1991

A fundamental assumption in conventional linear predictive coding (LPC) analysis procedure is that the input to an all-pole vocal tract filter is white process. In the case of periodic inputs, however, a pitch bias error is introduced into the conventional LP coefficient. Multi-pulse (MP) LP analysis can reduce this bias, provided that an estimate of the excitation is available. Since the prediction error of conventional LP analysis can be modeled as the sum of an MP excitation sequence and a random noise sequence, we can view extracting MP sequences from the prediction error as a classical detection and estimation problem. In this paper, we propose an algorithm in which the locations and amplitudes of the MP sequences are first obtained by applying a likelihood ratio test (LRT) to the prediction error, and LP coefficients free of pitch bias are then obtained from the MP sequences. To verify the performance enhancement, we iterate the above procedure with adaptive threshold at each step.

• Korean Journal of Environmental Biology
• /
• v.28 no.3
• /
• pp.143-149
• /
• 2010

The pathogens and community structure of the fishes in Taehwa river were investigated from March 2007 to January 2009. During the study period, 3,504 individuals belonging to 35 species, 17 families and 9 orders were collected. The numerically dominant and subdominat species were Opsarichthys uncirostris (relative abundance 39.7%) and Hemibarbus labeo (relative abundance 30.9%). There were five Korean endemic species (20.8%) including Squalidus chankaensis tsuchigae, Zacco koreanus, Cobitis hankugensis, Coreoperca herzi, Odontobutis platycephala. The large fishes like Hemibarbus labeo or Opsarichthys uncirostris were gathered around the Samho bridge, sampling site 2 according to a season. The reaction to which two kinds of fish pathogenic virus is all negative and no fish pathogenic bacteria was isolated from 220 individuals. The fish pathogenic parasite not present variously with 7 species. Especially, Trichodina sp. was detected monthly and the infective density was high. But it is cosidered that temporary overcrowding of fish is not influenced mass mortality causing diseases in the specific site of river.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.40 no.4
• /
• pp.346-353
• /
• 2012

The Laser-Induced Breakdown Spectroscopy (LIBS) has great advantages as an analytical technique, namely real-time analysis without sample preparation, ideal for mobile chemical sensor for space exploration. The LIBS plasma characteristics are strongly dependent on the surrounding pressure. In this study, seven types of target (C, Ti, Ni, Cu, Sn, Al, Zn) were investigated for their elemental lifetime. The target was located in vacuum chamber which has the pressure range of 760 to $10^{-5}$ torr. As the pressure is decreased, the elemental lifetimes of carbon and titanium declined, while all other targets showed increased lifetimes until reaching 1 torr and declined with continued pressure decrease. The boiling point and electronegativity amongst the physicochemical properties of the samples are used to explain this peculiarity.

• Journal of the Korean Society of Food Science and Nutrition
• /
• v.31 no.2
• /
• pp.338-342
• /
• 2002

Antimicrobial activity of yellow pigment produced by Monascus anka Y7 (Y7) was studied. The crude yellow pigment of Y7 showed antimicrobial activity against some bacteria and yeasts. The diameter of inhibition zone against gram-positive bacteria was a little smaller t]fan that of gram-negative bacteria to the crude yellow pigment. Especially, E2 fraction obtained from the crude yellow pigment by TLC method showed high anti-microbial activity against E. coli.. The fraction had bright yellow pigment, showing fluorescent light and having the maximum absorption at 373 nm. Citrinin, a mycotoxin which had been characterized as an antimicrobial substance from a Monascus strain, was not detected in the E2 fraction and in the crude yellow pigment by the results of TLC and HPLC. This indicates that the antimicrobial activity of Y7 pigments did not any relationship with citrinin. Yellow degree (b/a of Hunters color value) of Y7 pigment was much higher than that of other natural colorants such as annatto, gardenia yellow and carthamus yellow. But the colors of all of the yellow pigments were similar by panels. Thus, the yellow pigment of Y7 could be used as a useful alternative colorant for food industry, having the advantage of antimicrobial activity.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.12 no.1
• /
• pp.91-96
• /
• 2012

Ad-hoc networks can be used various environment, which it is difficult to construct infrastructures, such as shadowing areas, disaster areas, war area, and so on. In order to support to considerable and various wireless services, more spectrum resources are needed. However, efficient utilization of the frequency resource is difficult because of spectrum scarcity and the conventional frequency regulation. Ad-hoc networks employing cognitive radio (CR) system that guarantee high spectrum utilization provide effective way to increase the network capacity. In CR based wireless ad-hoc networks, cluster-head decides the existence of primary user using sensing information of primary user from each ad-hoc device. However, it is still defective research to decide cluster head among the a lot of ad-hoc devices. So, in this paper, we show the decision method of cluster head in CR based wireless and detection probabilities of primary user based on decision method of cluster head.

• The Journal of the Acoustical Society of Korea
• /
• v.35 no.6
• /
• pp.501-509
• /
• 2016

We propose a method to improve the robustness of speaker verification on short test utterances. The accuracy of the state-of-the-art i-vector/probabilistic linear discriminant analysis systems can be degraded when testing utterance durations are short. The proposed method compensates for utterance variations of short test feature vectors using deep neural networks. We design three different types of DNN (Deep Neural Network) structures which are trained with different target output vectors. Each DNN is trained to minimize the discrepancy between the feed-forwarded output of a given short utterance feature and its original long utterance feature. We use short 2-10 s condition of the NIST (National Institute of Standards Technology, U.S.) 2008 SRE (Speaker Recognition Evaluation) corpus to evaluate the method. The experimental results show that the proposed method reduces the minimum detection cost relative to the baseline system.