• 방송공학회논문지
• /
• 제24권2호
• /
• pp.227-233
• /
• 2019

본 논문은 가상현실 음향 구현을 위한 심층신경망 기반 사운드 보간 방법에 관한 것으로서, 이를 통해 두 지점에서 취득한 음향 신호들을 사용하여 두 지점 사이의 음향을 생성한다. 산술평균이나 기하평균 같은 통계적 방법으로 사운드 보간을 수행할 수 있지만 이는 실제 비선형 음향 특성을 반영하기에 미흡하다. 이러한 문제를 해결하기 위해서 본 연구에서는 두 지점과 목표 지점의 음향신호를 기반으로 심층신경망을 훈련하여 사운드 보간을 시도하였으며, 실험결과 통계적 방법에 비해 심층신경망 기반 사운드 보간 방법의 성능이 우수함을 보였다.

• International Journal of Advanced Culture Technology
• /
• 제7권2호
• /
• pp.218-226
• /
• 2019

In modern society, people live in an environment with artificial or natural noise. Especially, the sound that corresponds to the artificial noise makes the noise itself and affects each other because many people live and work in the city. Sounds are generated by the activities and causes of various people, such as construction sites, aircraft, production machinery, or road traffic. These sounds are essential elements in human life and are recognized and judged by human auditory organs. Noise is a sound that you do not want to hear by subjective evaluation, and it is a loud sound that gives hearing damage or a sound that causes physical and mental harm. In this study, we introduce the method of stimulating the human hearing by finger vibration and explain the advantages of the proposed method in various kinds of a noise environment. And how to improve the sound quality to improve efficiency. In this paper, we propose a method to prevent the loss of hearing loss and the transmission of sound information based on proper signal to noise ratio when using portable IT equipment in various noise environments.

• International Journal of Advanced Culture Technology
• /
• 제8권4호
• /
• pp.287-293
• /
• 2020

More and more people are paying attention to the psychological pleasure and relaxation that sound hearing brings. In most cases, humans seem to have a special preference for natural sounds. Natural sounds are mainly white noise and pink noise such as wind, rain, waves, waterfall sounds, etc. All of these are often considered to be beneficial to human health, but in reality the same category of natural sounds is no different. It will be very different due to space, time and other factors. Each sound can be unique, so people's hearing experience is also different. This paper quantitatively analyzes the spectrum and brain waves to analyze the feeling of hearing the natural Broken Falls sound. In particular, we aim to objectively analyze the objective feeling of Broken Falls sound falling on the human auditory system through sound spectrum and brain waves.

• Journal of the Korean Wood Science and Technology
• /
• 제32권6호
• /
• pp.50-56
• /
• 2004

This study was conducted to investigate the performance of sound absorption type-noise barriers manufactured with a combination of wood particles used for particleboard, recycled waste newspaper, and cement. An average density of wood-combined cement board was in the range from 0.83 to 0.96 g/cm³, showing relatively low-density board. Regardless of types of cement bonded board or wooden board, the board with concave holes(凹)-formed surfaces showed greater sound absorption coefficient compared to those of flat surface boards. The board density was not related with those coefficients. Accordingly, it was concluded that concave or deep corrugated surface structure has played an important role in sound absorption for the application of sound absorption type-noise barrier.

• 한국소음진동공학회논문집
• /
• 제16권5호
• /
• pp.457-469
• /
• 2006

In automotive engineering, the brand sound is one of the important advantage strategies in a car company. For the design of brand sound, the selection of descriptive word for a car sound is one of major works in automotive sound quality research. In this paper, booming and rumbling sound, which are professional words used by sound and vibration engineers are used for the design of brand sound. We employed sound quality metrics, which are used in the psychoacoustics. By most research results, the relationship between subjective evaluations and sound quality metrics has nonlinear characteristics. In order to correlate these subjective evaluations with sound quality metrics, the artificial neural network technology has been applied to two-dimensional sound quality index for a passenger car. These indexes are used for 46 passenger cars, which are samples of the famous cars around the world. Also a preference evaluation for car sound was carried out by sound and vibration engineers. We coupled this preference with booming and rumbling sounds by using artificial neural network. In future, the two dimensional sound and preference index will be very useful to develop brand sound in passenger cars.

• International Journal of Internet, Broadcasting and Communication
• /
• 제11권3호
• /
• pp.106-110
• /
• 2019

The interaural level difference (ILD) used for the sound localization using stereo signals is to find the difference in energy that the sound source reaches both ears. The conventional ILD does not consider the time difference of the stereo signals, which is a factor of lowering the accuracy. In this paper, we propose a synchronized ILD that obtains the ILD after synchronizing these time differences. This method uses the cross-correlation function (CCF) to calculate the time difference to reach both ears and use it to obtain synchronized ILD. In order to prove the performance of the proposed method, we conducted two sound localization experiments. In each experiment, the synchronized ILD and CCF or only the synchronized ILD were given as inputs of the deep neural networks (DNN), respectively. In this paper, we evaluate the performance of sound localization with mean error and accuracy of sound localization. Experimental results show that the proposed method has better performance than the conventional methods.

• Journal of the Korean Wood Science and Technology
• /
• 제52권3호
• /
• pp.234-242
• /
• 2024

In this study, the effect of carbonization temperature and mixing ratio of ceramics manufactured from sawdust, chaff and charcoal on sound absorption performance and density profile was investigated. The density profile of ceramics prepared by the addition rates of sawdust, chaff and charcoal showed the highest value at 91.00% when the ratio of sawdust, chaff and charcoal was 50:25:15. However, the difference in density profile according to the addition rate was insignificant. The density profile of ceramics manufactured according to the carbonization temperature showed the highest value of 88.06% when manufactured at 800℃. However, it does not show any particular trend, so it is understood that the effect of the carbonization temperature on the density gradient is small. On the other hand, the sound absorption coefficients of ceramics prepared by the addition rates of sawdust, chaff and charcoal is between 0.3 and 0.4 at almost all frequencies when the addition rates of sawdust, chaff and charcoal are 50:30:10 and 50:35:5, respectively. Therefore, as the chaff particles increased, the sound absorption performance was improved. In addition, the sound absorption coefficients of the ceramics manufactured at each carbonization temperature showed the highest value in the ceramics manufactured at 1,200℃.

• Journal of the Korean Wood Science and Technology
• /
• 제47권1호
• /
• pp.33-39
• /
• 2019

The square timbers of larch having cross section of $90mm{\times}90mm$ were glued laterally to be formed $1,200mm{\times}2,400mm$ panels which were used as cores for CLT wall panels. Then, structural plywood panels having size of $1,200mm{\times}2,400mm$ were used as cross band covering the small square timber cores to manufacture CLT wall panels. The sound absorption rate of CLT wall panels and polyester board attached CLT wall panels were investigated. The mean sound absorption coefficients of the former and the latter in the frequency range of 100-6400 Hz were 0.21 and 0.74, respectively. The noise reduction coefficients (NRC) of those were 0.21 and 0.40, respectively. Also, the mean sound transmission loss of CLT wood panel in the frequency range of 50-1600 Hz was 45.12 dB and that value at the frequency of 500 Hz was 42.49 dB. It was suggested that the polyester board attached CLT wall panels could be used as housing wall because of its high sound absorption rate and high sound transmission loss.

• 한국음향학회:학술대회논문집
• /
• 한국음향학회 1994년도 FIFTH WESTERN PACIFIC REGIONAL ACOUSTICS CONFERENCE SEOUL KOREA
• /
• pp.796-801
• /
• 1994

Measuring noise, sound quality or acoustical comfort presents a difficult task for the acoustic engineer. Sound and noise are ultimately jugded by human beings acting as analysers. Regulations for determining noise levels are based on A-weighted SPL measurement performed with only one microphone. This method of measurement is usually specified when determining whether the ear can be physically damaged. Such a simple measurement procedure is not able to determine annoyance of sound events or sound quality in general. For some years investigations with binaural measurement analysis technique have shown new possibilities for the objective determination of sound quality. By using Artificial Head technology /1/, /2/ in conjunction with psychoacoustic evaluation algorithms - and taking into account binaural signal processing of human hearing, considerable progress regarding the analysis of sounds has been made. Because sound events often arise in a complex way, direct conclusions about components subjectively judged to be annoying with regard to their causes and transmission paths, can be drawn in a limited way only. A new procedure, complementing binaural measurement technology combined with mulit-channel measuements of acceleration sensor signals has been developed. This involves correlating signals influencing sound quality, analyzed by means of human hearing, with signals form different acceleration sensors fixed at different positions of the sound source. Now it is possible to recognize the source and the transmission way of those signals which have an influence on the annoyance of sound.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2005년도 추계 학술대회논문집(수송기계편)
• /
• pp.174-179
• /
• 2005

In automotive engineering, the brand sound is one of the important advantage strategy in a car company. For the design of brand sound, the selection of descriptive word for a car sound is one of major works in automotive sound quality research. In paper, booming sound and rumbling sound, which are professional words used by NVH engineers are used for the design of brand sound. We employed sound metrics which are the subjective parameter used in psychoacoustics. According to most research results, the relationship between subjective evaluations and sound metrics has nonlinear characteristics and is very complex. In order to link these subjective evaluations to sound metrics, the artificial neural network technology has been applied to two-dimensional sound quality index for a passenger car. These indexes is used for 46 passenger cars, which are samples of famous cars in the world. Also the preference in car sounds is evaluated by the trained NVH engineers. We coupled this preference with booming and rumbling sounds by using artificial neural network. In future, the two -dimensional sound index and preference index are very useful fur the development of brand sound in passenger cars.