• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.13 no.10
• /
• pp.2211-2217
• /
• 2009

In this paper, we propose the speech enhancement algorithm using an one-microphone in a reverberant room environments. Spectral subtraction is the effective method which can reduce the reverberation element and the noise in a spectrum domain. Spectral subtraction needs correct separation of voice section and silent section therefore to improve the performance, voice activity detection(VAD) based on entropy has been applied to the proposed method. We test a performance of the proposed method by comparing with conventional method which used VAD based on energy detection. Reverberation reduction ratio with variable of SNR and a reverberation time is used as a test index. From the simulation result, proposed method shows performance better than conventional method.

• The Journal of the Acoustical Society of Korea
• /
• v.28 no.5
• /
• pp.447-453
• /
• 2009

This paper introduces the method for detect voices and exact end point at low SNR by maximizing voice energy. Conventional VAD (Voice Activity Detection) algorithm estimates noise level so it tends to detect the end point inaccurately. Moreover, because it uses relatively long analysis range for reflecting temporal change of noise, computing load too high for application. In this paper, the SEM-VAD (Speech Energy Maximization-Voice Activity Detection) method which uses psycho-acoustical bark scale filter banks to maximize voice energy within frames is introduced. Stable threshold values are obtained at various noise environments (SNR 15 dB, 10 dB, 5 dB, 0 dB). At the test for voice detection in car noisy environment, PHR (Pause Hit Rate) was 100%accurate at every noise environment, and FAR (False Alarm Rate) shows 0% at SNR15 dB and 10 dB, 5.6% at SNR5 dB and 9.5% at SNR0 dB.

• Journal of Korea Multimedia Society
• /
• v.17 no.9
• /
• pp.1064-1069
• /
• 2014

In this paper, we propose a novel approach to improve the performance of a voice activity detection(VAD) which is based on the adaptive band-partitioning with the likelihood ratio(LR). The previous method based on the adaptive band-partitioning use the weights that are derived from the variance of the spectral. In our VAD algorithm, the weights are derived from LR, and then the weights are incorporated with the entropy. The proposed algorithm discriminates the voice activity by comparing the weighted entropy with the adaptive threshold. Experimental results show that the proposed algorithm yields better results compared to the conventional VAD algorithms. Especially, the proposed algorithm shows superior improvement in non-stationary noise environments.

• Journal of rehabilitation welfare engineering & assistive technology
• /
• v.8 no.1
• /
• pp.41-46
• /
• 2014

In this paper, we propose a novel voice activity detection (VAD) algorithm using feature vectors based on TE (teager energy). Specifically, power spectral deviation (PSD), which is used as the feature for the VAD in the IS-127 noise suppression algorithm, is obtained after the input signal is transfomed by Teager energy operator. In addition, the TE-based likelihhod ratio are derived in each frame to modifiy the PSD for further VAD. The performance of our proposed VAD algorithm are evaluated by objective testing (total error rate, receiver operating characteristics, perceptual evaluation of speech quality) under various environments, and it is found that the proposed method yields better results than conventional VAD algorithms in the non-stationary noise environments under 5 dB SNR (total error rate = 2.6% decrease, PESQ score = 0.053 improvement).

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.38 no.12
• /
• pp.971-974
• /
• 2014

A ventricular assist device (VAD) is used for bridge to heart transplantation and heart diseases. Knowing the status of a pneumatic pulsatile VAD when implanting it into the body is important: when the velocity of blood flow through the VAD is slow, a thrombus may occur, and thrombosis can be fatal to a patient. In order to determine the state of a VAD, various sensors need to be implanted. Because this introduces the risk of infection and difficulties with sensor management, we developed a method for estimating the state of a VAD indirectly via the pressure in an air tube that can be measured in vitro. We compared the measured values to in vitro experimental results. The estimated and measured values showed some errors, but the accuracy can be improved by refining the estimation process to minimize the risk of infection.

• Journal of Chest Surgery
• /
• v.40 no.3 s.272
• /
• pp.168-179
• /
• 2007

Background: The recent trend of an increasing number of patients with acute cardiogenic shock or chronic congestive heart failure following myocardial infarction, as well as the considerable number who can not be weaned from cardiopulmonary bypass after open heart surgery, call for immediate efforts to develop affordable ventricular assist devices that are suitable for the Korean physique. Recently, a pneumatic pulsatile ventricular assist device (VAD), named DKUH-75, has been developed by the Department of Biomedical Engineering, in collaboration with the Department of Thoracic and Cardiovascular Surgery of Dankook University College of Medicine. The feasibility of the DKUH-75 VAD was evaluated on the bases of common hemodynamic variables and echocardiographic measurements in pigs, which are subjected to an acute cardiogenic shock state following myocardial infarction, using a novel coronary artery ligation method employing the ischemic preconditioning concept. Material and Method: Acute cardiogenic shock was induced in 10 Yorkshire Landrace Duroc strain pigs by ligating the left anterior descending coronary artery via an ischemic preconditioning process. The hemodynamic variables were monitored, with epicardial echocardiographic measurements performed before and one hour after the ligation. The DKUH-75 VAD was implanted into 5 pigs one hour after the onset of the shock. The hemodynamic variables and echocardiographic measurements were taken one hour after installation of the VAD. Result: The systolic, diastolic and mean systemic arterial pressures were significantly decreased in all the experimental animals one hour after the ligation. The systolic, diastolic and mean pulmonary arterial pressures were increased (Eds note: this completely contradicts the preceding statement? However, if you mean the non-experimental animals this should be stated?). The left ventricular end diastolic pressure (LVEDP) was increased, but the cardiac index decreased, An increase in the left ventricular end systolic dimension and decreases in the fractional shortening and ejection fraction were observed all animals one hour after the coronary artery ligation. In all 5 of the VAD implanted pigs, the systolic and mean systemic arterial pressures were increased, and the pulmonary arterial pressures decreased one hour after the implantation; the LVEDP decreased, but the cardiac index was significantly increased, In the echocardiographic measurements, the left ventricular end systolic dimension decreased after the implantation of the VAD, but the fractional shortening and ejection fraction significantly increased. Conclusion: Significant improvements in the hemodynamic variables and echocardiographic measurements were observed in the 5 VAD implanted animals one hour after installation, which had been subjected to an acute cardiogenic shock state by ligation of the coronary artery, indicating that the DKUH-75 VAD could help in the recovery of the myocardial function. This suggests that the DKUH-75 VAD is feasible in the short term in relation to an acute cardiogenic shock state due to myocardial infarction.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.45 no.4
• /
• pp.100-106
• /
• 2008

In this paper we apply a discriminative weight training employing power spectral flatness measure (PSFM) to a statistical model-based voice activity detection (VAD) in various noise environments. In our approach, the VAD decision rule is expressed as the geometric mean of optimally weighted likelihood ratio test (LRT) based on a minimum classification error (MCE) method which is different from the previous works in th at different weights are assigned to each frequency bin and noise environments depending on PSFM. According to the experimental results, the proposed approach is found to be effective for the statistical model-based VAD using the LRT.

• Journal of rehabilitation welfare engineering & assistive technology
• /
• v.8 no.3
• /
• pp.145-150
• /
• 2014

In this paper, we propose a novel scheme to improve the performance of a voice activity detection(VAD) which is based on the deep belief networks(DBN) with the likelihood ratio(LR). The proposed algorithm applies the DBN learning method which is trained in order to minimize the probability of detection error instead of the conventional decision rule using geometric mean. Experimental results show that the proposed algorithm yields better results compared to the conventional VAD algorithm in various noise environments.

• The Journal of the Acoustical Society of Korea
• /
• v.26 no.5
• /
• pp.194-198
• /
• 2007

In this paper, we apply a discriminative weight training to a statistical model-based voice activity detection(VAD). In our approach, the VAD decision rule is expressed as the geometric mean of optimally weighted likelihood ratios(LRs) based on a minimum classification error(MCE) method which is different from the previous works in that different weights are assigned to each frequency bin which is considered more realistic. According to the experimental results, the proposed approach is found to be effective for the statistical model-based VAD using the LR test.

• Journal of the Institute of Convergence Signal Processing
• /
• v.4 no.2
• /
• pp.18-24
• /
• 2003

One of the most important subjects in the implementation of real time speech recognizer is to design both reliable VAD(Voice Activity Detection) and suitable speech feature vector. But, because it is difficult to calculate reliable VAD in the environment having surrounding noise, designed suitable speech feature vector may not be obtained. Solving this problem, in this paper, we implement not only short time power spectrum which is generally used but also two additive parameters, the comparison measure of spectrum density having robust property in noise and linear discriminant function using linear regression, then perform VAD by using the combination of each parameter having apt weight in other magnitudes of surrounding noise and confirm that proposed parameters show a robust characteristic in circumstances having surrounding noise by using DTW(Dynamic Time Waning) in recognition experiment.