• The Journal of the Acoustical Society of Korea
• /
• v.21 no.5
• /
• pp.455-461
• /
• 2002

This paper presents a method that improves the performance of the personal voice dialling system in which speaker independent phoneme HMM's are used. Since the speaker independent phoneme HMM based voice dialing system uses only the phone transcription of the input sentence, the storage space could be reduced greatly. However, the performance of the system is worse than that of the system which uses the speaker dependent models due to the phone recognition errors generated when the speaker independent models are used. In order to solve this problem, a new method that jointly estimates transformation vectors for the speaker adaptation and transcriptions from training utterances is presented. The biases and transcriptions are estimated iteratively from the training data of each user with maximum likelihood approach to the stochastic matching using speaker-independent phone models. Experimental result shows that the proposed method is superior to the conventional method which used transcriptions only.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.50 no.11
• /
• pp.783-789
• /
• 2022

In a recent study, it was observed that the combustion gas entering the post chamber of a hybrid rocket contains vortices with very small size and high frequency characteristics. In addition, it was observed that small vortices collided with the nozzle wall to create a counter-flow, resulting in additional combustion with ignition delay. This study investigated the physical relationship between ignition delay induced by the counter-flow and the formation of beating pressure. To do this, a newly modified model was proposed by including ignition delay in the existing energy kicked oscillator model proposed by Culick. Numerical results show that the ignition delay is an important factor in determining the occurrence of the combustion pressure beats through the periodic formation of thermoacoustic coupling. In addition, when the ignition delay was reduced by increasing the post chamber length, the phase difference between the energy kick and the pressure generation was increased, the periodic pressure beats did not occur at all.

• Proceedings of the Acoustical Society of Korea Conference
• /
• autumn
• /
• pp.73-76
• /
• 2001

본 논문에서는 음소인식기와 음소결합확률모델을 이용하여 전화음성을 대상으로 입력음성이 어느 나라 말 인지를 식별할 수 있는 언어식별시스템을 구현하였고 성능을 실험하였다. 시스템은 음소인식기로 입력음성에 대한 음소열을 인식하는 과정, 인식된 음소열을 이용하여 인식대상 언어별 음소결합확률모델을 생성하는 훈련과정, 훈련과정에서 생성된 음소결합확률모델로부터 확률 값을 계산하여 인식결과를 출력하는 식별과정으로 구성된다. 본 논문에서는 음소결합확률모델로부터 우도를 계산할 때 정보이론(Information Theory, Shannon and Weaver, 1949)을 이용하여 가중치를 적용하는 방법을 제안하였다. 시스템의 훈련 및 실험에는 OGI 11개국어 전화음성 corpus (OGI-TS)를 사용하였으며, 음소인식기는 HTK를 이용하여 구현하였고 음소인식기 훈련에는 NTIMIT 전화음성 DB를 이용하였다. 실험결과 11개국어를 대상으로 45초 길이의 음성에 대해서 평균 $74.1\%$, 10초 길이의 음성에 대해서는 평균 $57.1\%$의 인식률을 얻을 수 있었다.

• Proceedings of the KSME Conference
• /
• 2008.11b
• /
• pp.2524-2529
• /
• 2008

The thermal system like a combustion chamber is believed to experience a significant instability problem with vibration in case that the thermal energy or the acoustic energy are transformed into a different form through a relevant path. This study deals with a numerically- predicted, Thermoacoustic instability in a Rijke tube by using a non-linear model for a heat source. The heating part where the energy transformation occurs actively is modeled after simulating two-dimensional cylinder case with constant surface temperature, and a nonlinear model that accounts for the transfer function of magnitude- and phase-characteristics is properly implemented so as to be dependent on the pulsation strength in the tube. The heat source model is observed to result in equivalent Thermoacoustic instabilities in the Rijke tube except low flow-rate cases in which the natural convection is dominant.

• The Journal of the Acoustical Society of Korea
• /
• v.42 no.4
• /
• pp.313-319
• /
• 2023

In this paper, a time series machine learning model, Long Short Term Memory (LSTM), is applied into the bubble flow noise data and the underwater projectile launch noise data to predict missing values of time-series underwater noise data. The former is mixed with bubble noise, flow noise, and fluid-induced interaction noise measured in a pipe and can be classified into three types. The latter is the noise generated when an underwater projectile is ejected from a launch tube and has a characteristic of instantaenous noise. For such types of noise, a data-driven model can be more useful than an analytical model. We constructed an LSTM model with given data and evaluated the model's performance based on the number of hidden units, the number of input sequences, and the decimation factor of signal. It is shown that the optimal LSTM model works well for new data of the same type.

• Journal of the Institute of Convergence Signal Processing
• /
• v.2 no.4
• /
• pp.9-15
• /
• 2001

In this paper we investigate context dependent acoustic models to improve the performance of the Korean speech recognition . The algorithm are using the Korean phonological rules and decision tree, By Successive State Splitting(SSS) algorithm the Hidden Merkov Netwwork(HM-Net) which is an efficient representation of phoneme-context-dependent HMMs, can be generated automatically SSS is powerful technique to design topologies of tied-state HMMs but it doesn't treat unknown contexts in the training phoneme contexts environment adequately In addition it has some problem in the procedure of the contextual domain. In this paper we adopt a new state-clustering algorithm of SSS, called Phonetic Decision Tree-based SSS (PDT-SSS) which includes contexts splits based on the Korean phonological rules. This method combines advantages of both the decision tree clustering and SSS, and can generated highly accurate HM-Net that can express any contexts To verify the effectiveness of the adopted methods. the experiments are carried out using KLE 452 word database and YNU 200 sentence database. Through the Korean phoneme word and sentence recognition experiments. we proved that the new state-clustering algorithm produce better phoneme, word and continuous speech recognition accuracy than the conventional HMMs.

• The Journal of the Acoustical Society of Korea
• /
• v.34 no.3
• /
• pp.210-218
• /
• 2015

When the driving voltage of an ultrasound transducer is increased to improve the quality of ultrasound images, heat is generated inside the transducer that can cause patient's skin burn and degradation of transducer performance. Hence, in this paper, a method to disperse the heat of the transducer has been studied. The phased array transducer having 3 MHz center frequency and 32 channels was selected for analyses of the thermal dispersion. First, mechanism of the heat generation was investigated in relation to the transducer operation through theoretical analysis, and material damping and sound pressure amplitude were confirmed to be influential on the heat generation. Further, we investigated the effects of the properties of the materials constituting the transducer on the thermal dispersion through finite element analysis. Based on the analysis results, we determined the thermal properties of the constituent materials that could facilitate the thermal dispersion inside the transducer. The determined thermal properties were applied to the finite element model, and the results showed that the maximum temperature at an acoustic lens contacting with a patient was decreased to 51 % of its initial value.

• Journal of the Korean Society of Propulsion Engineers
• /
• v.16 no.1
• /
• pp.1-8
• /
• 2012

Acoustic-pressure response of diluted hydrogen-air diffusion flames is investigated numerically by adopting a fully unsteady analysis of flame structures in low and high pressure regimes. As acoustic frequency increases, finite-rate chemistry is enhanced through a nonlinear accumulation of heat release rate for any pressure regime, leading to a high amplification index. Same numerical results are analyzed with application of a pressure-sensitive time lag model, and thereby, interaction index and time lag are calculated for each pressure regime. The interaction index has the largest value in each pressure regime at an acoustic frequency near 1000 Hz. In a high-pressure regime, flames are more unstable than in a low-pressure regime. The interaction index shows a good agreement with the amplification index.

• The Journal of the Acoustical Society of Korea
• /
• v.38 no.1
• /
• pp.96-105
• /
• 2019

Ocean reverberation is the most limiting factor in designing realistic and real-time system for sonar simulator. The simulation for an ocean reverberation requires a lot of computational loads, so it is hard to embed program and generate real-time signal in the sonar simulator. In this study, we simulate a time-domain bottom reverberation signal based on Harrison's energy-flux bottom reverberation model by applying Doppler effects as ship maneuvering and autoregressive model. Finally, the bottom reverberation signal with realistic characteristics could be generated for the simulation of ONR reverberation modeling workshop-I problem XI and East Sea ocean environments.

• The Journal of the Acoustical Society of Korea
• /
• v.5 no.4
• /
• pp.46-56
• /
• 1986

Hyperthermia를 사용하여 종양세포를 치료하는데 있어서 시간에 따른 동적 온도분포를 추정하 고 또, 그 온도분포에 관계하는 인자들이 동적 온도특성에 미치는 영향을 살펴보는 것은 실제 치료시에 정확한 온도제어를 위해서 반드시 필요하다. 본 논문에서는 몸속 10cm 깊이에 존재하는 원통형 종양모 델을 설정하고 초음파 동심환 변환기로 열을 집속시켰을 때 동적 온도분포 추정을 위해서 2차원 유한요 소법과 유한차분법을 이용하였다. 결과로서, 동적 온도분포에 가장 큰 영향을 미치는 인자는 혈류량이었 고, 추출된 동적 온도분포 특성값을 가지고 간단한 ON/OFF 온도제어에 적용할 수 있음을 보였다.