• 음성과학
• /
• 제9권1호
• /
• pp.49-59
• /
• 2002

This paper concerns an effective text-dependent speaker verification method to increase the performance of speaker verification. While various speaker verification methods have already been developed, their effectiveness has not yet been formally proven in terms of achieving acceptable performance levels. This paper proposes a weighted likelihood procedure along with a confidence measure based on subword-based text-dependent speaker verification. Our aim is to remedy the low performance problem in speaker verification by exploring a means to strengthen the verification likelihood via subword-based hypothesis criteria and weighted likelihood method. Experimental results show that the proposed speaker verification method outperforms that of the speaker verification scheme without using the proposed decision by a factor of up to 1.6 times. From these results, the proposed speaker verification method is shown to be very effective and to achieve a reliable performance.

• 음성과학
• /
• 제7권4호
• /
• pp.41-47
• /
• 2000

This paper concerns an effective text-prompted speaker verification method to increase the performance of speaker verification. While various speaker verification methods have already been developed, their effectiveness has not yet been formally proven in terms of achieving an acceptable performance level. It is also noted that the traditional methods were focused primarily on single, prompted utterance for verification. This paper, instead, proposes sequential decision method using variable threshold focused at handling two utterances for text-prompted speaker verification. Experimental results show that the proposed speaker verification method outperforms that of the speaker verification scheme without using the sequential decision by a factor of up to 3 times. From these results, we show that the proposed method is highly effective and achieves a reliable performance suitable for practical applications.

• 한국통신학회논문지
• /
• 제21권7호
• /
• pp.1642-1653
• /
• 1996

This paper covers feature extraction for automatic handwritten signature verification. Several major feature selection techniques are investigated from a practical perspective to realise an optimal signature verification system, and customised feature set selection based on set-on-set distance measurement is presented. The experimental results have proved the proposed methods to be efficient, offering considerably improved verification performance compared to conventional methods. Also, they dramatically reduce the processing complexity in the verification system.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2000년도 추계학술대회 논문집
• /
• pp.285-288
• /
• 2000

This paper deals with the theoretical estimation and its experimental verification of grinding wheel wear in surface grinding process. A theoretical formulation is provided to predict the grinding wheel wear in surface grinding. The associated surface roughness and grinding force are also investigated both theoretically and experimentally. Through a series of simulations and experiments, it is shown that the predictions are in good agreement with the experimental results.

• 대한기계학회논문집A
• /
• 제26권2호
• /
• pp.267-274
• /
• 2002

Using a flexible rotating beam test bed, experimental verification of a flexible multibody dynamic simulations for a rotating beam model has been carried out. The test bed consists of a flexible arm, harmonic driver reducer, AC servo motor and DSP board with PC. The mechanical ports of the test bed has been designed using 3D CAD program. For the simulation model, mass and moment of inertia of each part of the flexible rotating beam test bed are also obtained from 3D CAD model. In the flexible multibody dynamic simulations, the substructuring model has been established to capture nonlinear effects of the flexible rotating beam. Through the experimental verification, substructuring model provides better results than those from the linear model in the high speed rotation.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2009년도 제40회 하계학술대회
• /
• pp.1385_1386
• /
• 2009

In this paper, we examined the faulted cable joint through the external form analysis, material analysis, experimental verification and it's cause diagnosis system. It was not observed the voild, sharp material from the external form analysis and material variation. From the experimental verification, the thickness decrease of an insulator decreased ac breakdown strength suddenly and the breakdown traces of the insulator that was damaged by knife displayed elliptic shape. Thus, the faulted cable is assumed to accident that become dielectric breakdown by the deterioration of insulation performance that can happen when work.

• 전자공학회논문지CI
• /
• 제45권5호
• /
• pp.162-172
• /
• 2008

본 논문에서는 이동환경에서 개인의 신원을 인증하는 수단으로 치열영상과 음성을 생체정보로 이용한 멀티모달 화자인증 방법에 대하여 제안한다. 제안한 방법은 이동환경의 단말장치중의 하나인 스마트폰의 영상 및 음성 입력장치를 이용하여 생체 정보를 획득하고, 이를 이용하여 사용자 인증을 수행한다. 더불어, 제안한 방법은 전체적인 사용자 인증 성능의 향상을 위하여 두 개의 단일 생체인식 결과를 결합하는 멀티모달 방식으로 구성하였고, 결합 방법으로는 시스템의 제한된 리소스를 고려하여 비교적 간단하면서도 우수한 성능을 보이는 가중치 합의 방법을 사용하였다. 제안한 멀티모달 화자인증 시스템의 성능평가는 스마트폰에서 획득한 40명의 사용자에 대한 데이터베이스를 이용하였고, 실험 결과, 치열영상과 음성을 이용한 단일 생체인증 결과는 각각 8.59%와 11.73%의 EER를 보였으며, 멀티모달 화자인증 결과는 4.05%의 EER를 나타냈다. 이로부터 본 논문에서는 인증 성능을 향상하기 위하여 두 개의 단일 생체인증 결과를 간단한 가중치 합으로 결합한 결과, 높은 인증 성능의 향상을 도모할 수 있었다.

• The Korean Journal of Physiology and Pharmacology
• /
• 제28권1호
• /
• pp.59-72
• /
• 2024

To investigate the mechanism of Wenshen Xuanbi Decoction (WSXB) in treating osteoarthritis (OA) via network pharmacology, bioinformatics analysis, and experimental verification. The active components and prediction targets of WSXB were obtained from the TCMSP database and Swiss Target Prediction website, respectively. OA-related genes were retrieved from GeneCards and OMIM databases. Protein-protein interaction and functional enrichment analyses were performed, resulting in the construction of the Herb-Component-Target network. In addition, differential genes of OA were obtained from the GEO database to verify the potential mechanism of WSXB in OA treatment. Subsequently, potential active components were subjected to molecular verification with the hub targets. Finally, we selected the most crucial hub targets and pathways for experimental verification in vitro. The active components in the study included quercetin, linolenic acid, methyl linoleate, isobergapten, and beta-sitosterol. AKT1, tumor necrosis factor (TNF), interleukin (IL)-6, GAPDH, and CTNNB1 were identified as the most crucial hub targets. Molecular docking revealed that the active components and hub targets exhibited strong binding energy. Experimental verification demonstrated that the mRNA and protein expression levels of IL-6, IL-17, and TNF in the WSXB group were lower than those in the KOA group (p < 0.05). WSXB exhibits a chondroprotective effect on OA and delays disease progression. The mechanism is potentially related to the suppression of IL-17 and TNF signaling pathways and the down-regulation of IL-6.

• 음성과학
• /
• 제7권1호
• /
• pp.31-37
• /
• 2000

This paper concerns effective speaker adaptation methods to solve the over-training problem in speaker verification, which frequently occurs when modeling a speaker with sparse training data. While various speaker adaptations have already been applied to speech recognition, these methods have not yet been formally considered in speaker verification. This paper proposes speaker adaptation methods using a combination of MAP and MLLR adaptations, which are successfully used in speech recognition, and applies to speaker verification. Experimental results show that the speaker verification system using a weighted MAP and MLLR adaptation outperforms that of the conventional speaker models without adaptation by a factor of up to 5 times. From these results, we show that the speaker adaptation method achieves significantly better performance even when only small training data is available for speaker verification.

• Journal of Information Processing Systems
• /
• 제14권4호
• /
• pp.807-823
• /
• 2018

Speaker verification system performance depends on the utterance of each speaker. To verify the speaker, important information has to be captured from the utterance. Nowadays under the constraints of limited data, speaker verification has become a challenging task. The testing and training data are in terms of few seconds in limited data. The feature vectors extracted from single frame size and rate (SFSR) analysis is not sufficient for training and testing speakers in speaker verification. This leads to poor speaker modeling during training and may not provide good decision during testing. The problem is to be resolved by increasing feature vectors of training and testing data to the same duration. For that we are using multiple frame size (MFS), multiple frame rate (MFR), and multiple frame size and rate (MFSR) analysis techniques for speaker verification under limited data condition. These analysis techniques relatively extract more feature vector during training and testing and develop improved modeling and testing for limited data. To demonstrate this we have used mel-frequency cepstral coefficients (MFCC) and linear prediction cepstral coefficients (LPCC) as feature. Gaussian mixture model (GMM) and GMM-universal background model (GMM-UBM) are used for modeling the speaker. The database used is NIST-2003. The experimental results indicate that, improved performance of MFS, MFR, and MFSR analysis radically better compared with SFSR analysis. The experimental results show that LPCC based MFSR analysis perform better compared to other analysis techniques and feature extraction techniques.