• The Journal of the Acoustical Society of Korea
• /
• v.41 no.3
• /
• pp.335-341
• /
• 2022

In this paper, we adopt a transformer that shows remarkable performance in natural language processing as an acoustic model of hybrid speech recognition. The transformer acoustic model uses attention structures to process sequential data and shows high performance with low computational cost. This paper proposes a method to improve the performance of transformer AM by applying each of the four algorithms of sequence discriminative training, a weighted finite-state transducer (wFST)-based learning used in the existing DNN-HMM model. In addition, compared to the Cross Entropy (CE) learning method, sequence discriminative method shows 5 % of the relative Word Error Rate (WER).

• Journal of the Korea Institute of Military Science and Technology
• /
• v.27 no.3
• /
• pp.416-427
• /
• 2024

The wake-homing underwater guided weapon that detects and tracks wake generated during voyage of a surface ship is impossible to avoid with the present acoustic deception torpedo defense system. Therefore, research on bubble-generating wake decoy is necessary to deceive wake-homing underwater guided weapon. Experiments in various environments are required to verify the effective operation method and performance of the wake decoy, but performance verification through underwater experiment is limited. In this paper, we develop a simulator for an torpedo defense system of surface ship, which is applied bubble-generating wake decoy, against acoustic, wake, and hybrid homing underwater guided weapon attack. The simulator includes surface ship model, acoustic decoy(static, mobile) model, bubble-generating wake decoy model, search and motion model of underwater guided weapon and so on. By integrating various models, MATLAB GUI simulator was developed. Through the simulation results for various environmental variables by this simulator, it is judged that effective operation method and performance verification of the bubble-generating wake decoy can be performed.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2011.06a
• /
• pp.1057-1058
• /
• 2011

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.27 no.8
• /
• pp.1165-1173
• /
• 2003

The effects of (1) phonon dispersion on thermal conductivity model and (2) differentiation of group velocity and phase velocity are examined for germanium. The results show drastic change of thermal conductivity regardless of the same relaxation time model. Also the contribution of transverse acoustic (TA) phonon and longitudinal acoustic (LA) phonon on the thermal conductivity at high temperatures is reassessed by considering more rigorous dispersion model. Holland model, which is commonly used for modeling thermal conductivity, underestimates the scattering rate for TA phonon at high frequency. This leads the conclusion that TA is dominant heat transfer mode at high temperatures. But according to the rigorous consideration of phonon dispersion, the reduction of thermal conductivity is much larger than the estimation of Holland model, thus the TA at high frequency is expected to be no more dominant heat transfer mode. Another heat transfer mechanism may exist at high temperatures. Two possible explanations we the roles of (1) Umklapp scattering of LA phonon at high frequency and (2) optical phonon.

• Proceedings of the KSME Conference
• /
• 2003.04a
• /
• pp.1627-1632
• /
• 2003

The effect of (1) phonon dispersion in thermal conductivity model and (2) the differentiation of group velocity and phase velocity for Ge is examined. The results show drastic change of thermal conductivity regardless of using same relaxation time model. Also the contribution of transverse acoustic (TA) phonon and longitudinal acoustic (LA) phonon is changed by considering more rigorous dispersion model. Holland model underestimates the scattering rate for high frequency TA, so misleading conclusion, i.e. TA is dominant heat transfer mode at high temperature. But the actual reduction of thermal conductivity is much larger than the estimation by Holland model and high frequency TA is no more dominant heat transfer mode. Another heat transfer mechanism may exist for high temperature. Two possible explanations are (1) high frequency LA by Umklapp scattering and (2) optical phonon.

• MALSORI
• /
• no.43
• /
• pp.137-150
• /
• 2002

Speech segmentation at phoneme level is important for corpus-based text-to-speech synthesis. In this paper, we examine acoustic modeling methods to improve the performance of automatic speech segmentation system based on Hidden Markov Model (HMM). We compare monophone and triphone models, and evaluate several model training approaches. In addition, we employ an energy-based postprocessing scheme to make correction of frequent boundary location errors between silence and speech sounds. Experimental results show that our system provides 71.3% and 84.2% correct boundary locations given tolerance of 10 ms and 20 ms, respectively.

• The Journal of the Acoustical Society of Korea
• /
• v.20 no.3E
• /
• pp.10-17
• /
• 2001

The possibility is investigated of determining the diffuse absorption coefficients of the wall surfaces in a real room by minimizing the errors between the measured energy impulse response of a real room and the predicted energy impulse responses obtained from the ray tracing simulation of the room. In other words, this can possibly serve as a basis for "acoustical system identification" in attempting to determine the "best fit" of modelled absorption coefficients to measured energy response data. Algorithms for attempting this were investigated. The algorithms developed for this purpose proved to be rigorous and efficient. Instead of using the ray tracing model to determine the absorption coefficients, the phase image model was used in order to determine the acoustic impedances of wall surfaces. However, the numerical algorithms could not find the correct impedance values, primarily due to the wide range of the acoustic impedance values of any single acoustic material and very long computation time.

• Journal of The Korean Astronomical Society
• /
• v.18 no.1
• /
• pp.15-31
• /
• 1985

Making use of the arbitrary shock theory developed by Ulmschneider (1967, 1971) and Ulmschneider and Kalkofen (1978), we have calculated the dissipation rates of upward-travelling slow-mode acoustic shock waves in umbral chromospheres for two umbral chromosphere models, a plateau model by Avrett (1981) and a gradient model by Yun and Beebe (1984). The computed shock dissipation rates are compared with the radiative cooling rate given by Avrett (1981). The results show that the slow-mode acoustic shock waves with a period of about 20 second can heat the low umbral chromospheres travelling with a mechanical energy flux of $2.6{\times}10^6\;erg/cm^2s$ at a height of $300{\sim}400km$ above the temperature minimum region.

• Proceedings of the KIEE Conference
• /
• 2011.07a
• /
• pp.1788-1789
• /
• 2011

In order to develop a one-axis gimbaled acoustic source tracker for mobile robots, a pseudo-linear direction of arrival(DOA) estimator is proposed using a linear ultrasonic sensor array. Under the assumption that the sensor measurement errors are negligible, a linear measurement model is derived using the linear prediction relation of the received sinusoidal acoustic signals. Applying the Kalman filtering technique for this model, the linear recursive DOA estimator is designed. For its linear recursive filter structure, it is preferable for real-time implementation on a commercial DSP. Through the experiments, the effectiveness of the suggested method is demonstrated.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.14 no.1
• /
• pp.55-61
• /
• 2011

This paper presents bubble wake measurement results generated by the planing hull. The bubble was generated by SNAME TMB model(No. 4876) with hard chine at the CWC(Circulating Water Channel). ABS(Acoustic Bubble Spectrometer) was used to measure bubble wake measurement. The manufactured model is one meter in length and uniform velocity to generate the bubble at CWC is 3m/s, relatively higher speed than conventional hull form. Measurements were performed successfully and measured results show well the general characteristics of bubble wake generated by planing hull. Furthermore, experimental equations are proposed for the practical use.