• 한국음향학회지
• /
• 제21권4호
• /
• pp.146-146
• /
• 2002

We present a new technique for achieving source separation when given only a single channel recording. The main idea is based on exploiting the inherent time structure of sound sources by learning a priori sets of time-domain basis functions that encode the sources in a statistically efficient manner. We derive a learning algorithm using a maximum likelihood approach given the observed single channel data and sets of basis functions. For each time point we infer the source parameters and their contribution factors. This inference is possible due to the prior knowledge of the basis functions and the associated coefficient densities. A flexible model for density estimation allows accurate modeling of the observation, and our experimental results exhibit a high level of separation performance for simulated mixtures as well as real environment recordings employing mixtures of two different sources. We show separation results of two music signals as well as the separation of two voice signals.

• 한국컴퓨터정보학회논문지
• /
• 제12권1호
• /
• pp.51-56
• /
• 2007

음악을 분류하기 위해 특성함수를 사용하여 추출한 특성값 벡터를 사용한다. 본 실험에서는 특성값 벡터를 추출하기 위해 스펙트랄 롤오프, 분산, 평균 피크레벨을 사용하였다. 이중에서 스펙트랄 롤오프는 저음프레임과 고음프레임의 상대적인 비를 나타낸다. 최적의 롤오프 포인트를 찾기 위하여 롤오프 포인트를 0.05에서 0.9까지 0.05간격으로 증가시키며 반복실험 하였다. 롤오프 포인트를 증가시키며 분류성공률을 관찰하였다. 그리고 실험에 사용된 음원데이터는 바로크바이올린과 현대바이올린 연주이다. 두 종류의 악기는 모양과 주파수대역에 있어서 유사하지만 약간의 대역차와 질감의 차이를 가지고 있다. 이러한 특성이 최적의 롤오프 포인트를 찾는데 유용할 것으로 판단하였다. 실험결과 롤오프 포인트 0.85에서 가장 높은 분류성공률 85%를 나타냈다.

• 한국음향학회지
• /
• 제29권3호
• /
• pp.173-183
• /
• 2010

본 논문에서는 원통형 공동계의 음향 전달 특성에 대하여 이론적으로 검토한다. 원통형 공동계는 성덕대왕 신종의 음향 공동을 단순화시킨 모델로, 주 공동, 보조 공동 및 간극으로 구성되어 있다. 주 공동 내부에 점음원을 주고 내부 및 외부 음장에서의 주파수 응답 특성과 음향 모드를 구하였다. 그 결과는 SYSNOISE를 이용한 경계요소 해석결과와 비교하여 타당성을 확인하였다. 제안된 이론모델을 이용하여 간극과 보조 공동이 음향계의 공명 주파수와 음향 전달 특성에 미치는 영향을 규명하였다. 마지막으로 음원 주파수를 공동 내부에서 외부로 가장 잘 전달시키는 보조 공동과 간극의 조합을 결정하였다.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2000년도 춘계학술대회논문집
• /
• pp.319-324
• /
• 2000

Sound insulation effect of periodically arranged circular rods is studied by numerical method (BEM) and model experiment in anechoic room. Five different cylinder groups of diameter from 48 mm to 27 mm are arranged, where each group consists of 5 rows of identical rods and area ratio is kept almost the same as 60%. Comparisons of results by BEM and experiment show the band gap frequency ranges are qualitatively in agreement, while experiments gives higher attenuations. Frequency shifted attenuations (BEM results) by first group (${\phi}$48), third group (${\phi}$34) and fifth group (${\phi}$27) show that attenuation curves are in reasonable agreements for both plane wave and point source.

• 한국소음진동공학회논문집
• /
• 제25권8호
• /
• pp.539-545
• /
• 2015

Intense acoustic load is generated when a launch vehicle lifts off, causing the damaging vibrations at the launch vehicle or satellite within the fairing. This paper is concerned with the prediction of lift-off acoustic loads for a launch vehicle. As a test example, the lift-off acoustic load on the Korean launch vehicle, NARO, is predicted by the existing calculation tool, the modified Eldred's second method. Although the acoustic sources, assumed as point sources, are to be located along the center line of the exhaust plume when using the Eldred's prediction method, the exact location of the deflected center line of exhaust gas flow is not usually known. To search for the most appropriate source positions, six models of source line distribution are suggested and the acoustic load prediction results from these models are compared with the actual measurements. It is found that the predicted sound pressure spectrum of the Naro is the most similar to the measured data when the centerline of the turbulent kinetic energy contour is used as the source line.

• 한국안전학회지
• /
• 제11권4호
• /
• pp.135-142
• /
• 1996

Prolonged in-plant personnel exposure to high noise levels results in permant hearing damage. There are no way to correct this hearing damage by treatment or use of hearing aids. Therefore, every employer is responsible for providing a workplace free of such hazards as excessive noise. This study was carried out to evalute and predict a given noise environment based on specific limit as the noise guarantee for a newly-founded petrochemical plant. The maximum total sound level should not exceed 85dBA in the work area, except where the area is defined as a restricted area and 70dBA at the plant boundary. Prediction of the noise levels within the plant area for a newly-founded petrochemical plant was achieved by dividing all plant area into 20m$\times$20m regular grid spaces and noise level inside the area or unit that in-plant personel exposure to high noise levels was estimated computed into 5m$\times$5m regular grid spaces. The noise level at the grid point that was propagated from each of the noise sources(equipments) computed using the methematical formula was defined as follows : $SPL_2$=$SPL_1-20log{\frac{r_2}{r_1}}$(dB) where $SPL_1$ =sound pressure level at distance $r_1$ from the source $SPL_2$=sound pressure level at distance $r_2$ from the source As a result, the equipments exceeded noise limit or irritaring noise levels were identified on the specific grid coordinates. As for equipments in the area that show high noise levels, appropriate counter-measures for noise control (by barriers, enclosure, silencers, or the change of equipments, for example) should be reviewed. Methods for identifying sources of noise applied in this study should be the model for prediction of the noise levels for any newly-founded plant.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2000년도 하계종합학술대회 논문집(4)
• /
• pp.92-97
• /
• 2000

An AMR(Adaptive Multi-Rate) speech coding algorithm has been adopted as a standard speech codec for IMT-2000. It is based on the algebraic CELP, and consists of eight speech coding modes having the bit rate from 4.75 kbit/s to 12.2 kbit/s. It also contains the VAD(Voice Activity Detector), SCR (Source Controlled Rate) operation, and error concealment scheme for robustness in a radio channel. The bit rate of AMR is changed on a frame basis depending on the channel condition. In this paper, we introduced AMR speech coding algorithm and performed the real-time implementation using TMS320C6201, i.e., a Texas Instrument's fixed-point DSP. With the ANSI C source code released from ETSI and 3GPP, we convert and optimize the program to make it run in real time using the C compiler and assembly language. It is verified that the decoded result of the implemented speech codec on the DSP is identical with the PC simulation result using ANSI C code for test sequences. Also, actual sound input/output test using microphone and speaker demonstrates its proper real-time operation without distortions or delays.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2002년도 추계학술대회 논문집
• /
• pp.879-882
• /
• 2002

To control the vibration and sound of structure, it is important to analyze the dynamic action of structure. And through those analysis, the vibration source and the flow path is understood. To grasp that, when the two-dimension plate structure is shaken by a harmonic point excitation with the natural frequency using the finite element method, this paper presents the relation between vibration power flow and mode shape. As those results present to vector flow, the vibration power flow is visualized.

• 한국음향학회:학술대회논문집
• /
• 한국음향학회 1994년도 FIFTH WESTERN PACIFIC REGIONAL ACOUSTICS CONFERENCE SEOUL KOREA
• /
• pp.791-795
• /
• 1994

In this paper, the echo phenomena of Yingying Pagoda(ancient Chinese architecture), which may be resulted from interferences of reflection and diffraction by the pagoda eaves when pulse sound source is at some suitable positions, are investigated by an 1:2 scale model. There are valleys in frequency spectrum due to the interferences. On the other hand, taking eaves as wedges approximately, numerical spectral estimates are obtained from the closed-form impulse solution for diffraction of pulse point-source radiation by an infinite rigid wedge. The results of the numerical computations are similar to those of the model experiments. The study is a helpful guide to reconstruction or maintenance of this kind of ancient buildings.

• 대한기계학회논문집A
• /
• 제25권12호
• /
• pp.2100-2108
• /
• 2001

The impinging tones by high-speed plane jets are experimentally investigated to study the edgetone characteristics. Experiment used a slit nozzle and a wedge system to generate edgetones. The jet in the experiment is varied from low to high subsonic speed to obtain the effect of the speed on the frequency characteristics of impinging tones. The experimental data obtained previously fur edgetones and platetones by various nozzles are compared with the present edgetone data. And the condition of tone generation, the frequency ranges, the effective source point and the sound pressure level are compared and discussed. It is found that the jet speed has no diect influence on the impinging tone characteristics. Regardless of the jet speed, the effective source point is about a quarter wavelength downstream from the edge tip. With increase in jet speed, the influence of the nozzle configuration is decreased and the operating frequencies show good coincidences by normalized parameters based on the slit thickness.