• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.8
• /
• pp.365-370
• /
• 2023

This paper aims to compare the performance of speech data classification into two groups, adult and elderly, based on the acoustic and linguistic characteristics that change due to aging, such as changes in respiratory patterns, phonation, pitch, frequency, and language expression ability. For acoustic features we used attributes related to the frequency, amplitude, and spectrum of speech voices. As for linguistic features, we extracted hidden state vector representations containing contextual information from the transcription of speech utterances using KoBERT, a Korean pre-trained language model that has shown excellent performance in natural language processing tasks. The classification performance of each model trained based on acoustic and linguistic features was evaluated, and the F1 scores of each model for the two classes, adult and elderly, were examined after address the class imbalance problem by down-sampling. The experimental results showed that using linguistic features provided better performance for classifying adult and elderly than using acoustic features, and even when the class proportions were equal, the classification performance for adult was higher than that for elderly.

• The Journal of the Acoustical Society of Korea
• /
• v.43 no.3
• /
• pp.261-269
• /
• 2024

In this paper, the acoustic performance of an absorptive silencer was enhanced by optimizing an arrangement of multi-layered absorbing materials. The acoustic performance of the silencer was evaluated through transmission loss, and finite element method-based numerical analysis program was employed to calculate the transmission loss. Polyurethane, a porous elastic material frequently used in absorptive silencers, was employed as the absorbing material. The Biot-Allard model was applied, assuming that air is filled inside the polyurethane. By setting the frequency range of interest up to the 2 kHz and the acoustic performance affecting properties of the absorbing materials were investigated when it was composed as a single layer. And the acoustic performance of the silencers with the single and multi-layered absorbing materials was compared with each other based on polyurethane material properties. Subsequently, the arrangement of the absorbing materials was optimized by applying the Nelder-Mead method. The results demonstrated that the average transmission loss improved compared to the single-layered absorptive silencer.

• The Journal of the Acoustical Society of Korea
• /
• v.33 no.4
• /
• pp.225-231
• /
• 2014

This paper suggests an algorithm to improve the tracking performance for an underwater acoustic target using the feedback information of acoustic feature of a target. While conventional tracking algorithms use detected acoustic signals only, the proposed algorithm uses detected acoustic signals and target feature information as well. Since the proposed algorithm tracks only the selected measurements using target feature information, it prevents onset of unnecessary tracks and improves tracking performance for target of interest. Furthermore, it optimizes tracking parameters for the target of interest and guarantees robustness and consistency of the track. Some simulations are performed to demonstrate the improved tracking performance of the proposed algorithm.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.21 no.4
• /
• pp.357-364
• /
• 2011

In this study, ring-shaped acoustic resonators were proposed to be installed in a silencer to increase the acoustic performance of silencer in the low-frequency range. Evaluation of noise reduction performance of acoustic resonator arrays was carried out by measuring the random-incidence absorption coefficient. It was found that the absorption coefficient of resonator array was measured up to 1 at 125 Hz of 1/3-octave band center frequency. Insertion losses of silencers with ring-shaped acoustic resonator arrays were measured based on ISO 7235. The results were shown that the ring-shaped resonator could increase the insertion loss up to 13 dB without flow, whereas 7 dB when flow speed reached 15 m/s. As increasing the flow speed above 15 m/s, the effect of acoustic resonator decreased due to the effect of nonlinear air damping of the resonator. It was also found that the increment of pressure drop by the presence of resonator arrays was about 9 % at flow speed of 25 m/s.

• KIPS Transactions on Computer and Communication Systems
• /
• v.5 no.3
• /
• pp.71-78
• /
• 2016

Recently, The interest about underwater acoustic communication is increase such as marine resources, disaster prevention, weather prediction, and so on. Because the underwater acoustic communication uses a water as media, the underwater acoustic communication has a lot of restrictions. Although the underwater acoustic communication is hard, it is important to consider the security. In this paper, we estimate the performance of cryptographic algorithms(AES, ARIA, and LEA) on a various devices, available in underwater acoustic communication, and analysis the results. This result will be provide effective data confidentiality for underwater communication.

• International Journal of Naval Architecture and Ocean Engineering
• /
• v.6 no.1
• /
• pp.74-86
• /
• 2014

The optimization techniques are explored in the direction of arrival (DOA) estimation based on single acoustic pressure gradient vector sensor (APGVS). By analyzing the working principle and measurement errors of the APGVS, acoustic intensity approaches (AI) and the minimum variance distortionless response beamforming approach based on single APGVS (VMVDR) are deduced. The radius to wavelength ratio of the APGVS must be not bigger than 0.1 in the actual application, otherwise its DOA estimation performance will degrade significantly. To improve the robustness and estimation performance of the DOA estimation approaches based on single APGVS, two modified processing approaches based on single APGVS are presented. Simulation and lake trial results indicate that the performance of the modified approaches based on single APGVS are better than AI and VMVDR approaches based on single APGVS when the radius to wavelength ratio is not bigger than 0.1, and the two modified DOA estimation methods have excellent estimation performance when the radius to wavelength ratio is bigger than 0.1.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2007.11a
• /
• pp.236-241
• /
• 2007

Recently, in accordance with the cultural progress and the improvement of citizen's consciousness level, it is the real situation that the construction of small-scaled Multi-Purpose Hall where various cultural events could be performed, is on increasing. However, because the most of Multi-Purpose Hall had been designed and built up without any consideration on Acoustic Factor, many problems are on occurring thereat. Because those small-scaled Multi-Purpose Hall have been mostly used with the finishing material which contains a high degree of Acoustic Absorption indiscreetly, both diffusion and reflection of sound are not establishing properly, and since the reverberation of sound is very low, in case of musical performance by musical instrument, its sound hears too arid and stiff, there occurs some acoustic defect such as it becomes difficult for music appreciation with sufficient timbre, so that the capability improvement on the matter is urgently requiring situation. On this viewpoint, this Research has caught the Authenticity of the forecasted value and actually measured value using Acoustic Simulation on the subject of the small-scaled Multi-Purpose Hall, and then intended to propose the optimum acoustic condition on the basis of the Authenticity. As the result, it could be designed a Multi-Purpose Hall equipped with the satisfactory acoustic performance compare with the time before improvement, so that it is considered that such research result could be practically used as the fundamental material when design for other similar small-scaled Multi-Purpose Hall in the future.

• Proceeding of Spring/Autumn Annual Conference of KHA
• /
• 2008.04a
• /
• pp.421-426
• /
• 2008

In case of the large-scaled indoor gymnasium that has been constructed in the local area, there are many instances for the use as multipurpose space where the public performances are possible such as leisure activity, lecture, assembling activity, drama, concert and so on for the resident together with the purpose of sporting facility. In order for utilization to the maximum of the function of such indoor gymnasium, the acoustic capabilities concerned with Definition of both Voice and Music are simultaneously required. However, in case of the large-scaled athletic facility, since it was designed with high ceiling-height in view of its characteristics, it forms a Sound Focus and then the sound is concentrating to the specific part, and because the vibration of sound is too loud due to its broad volume, the acoustic defects such as many Echoes are arising. On such viewpoint, based on the drawing of the indoor gymnasium that is scheduled to be built at B County, Chonbuk Province, this Study has proposed such indoor gymnasium equipped with the optimized acoustic condition passing through Acoustic Simulation Phase. As the result of Acoustic Simulation, we could design the indoor gymnasium that equipped with really satisfying acoustic performance 'after' reformation compared with 'before' reformation, and it is considering that such material could be utilized as the fundamental material that brings a curtailment effect of the construction cost and also enables us to improve the acoustic performance, at the stage of planning and designing for the similar sporting facility in the future.

• Phonetics and Speech Sciences
• /
• v.10 no.1
• /
• pp.33-38
• /
• 2018

Although deep neural network-based acoustic models have greatly improved the performance of automatic speech recognition (ASR), reverberation still degrades the performance of distant speech recognition in indoor environments. In this paper, we adopt the DenseNet, which has shown great performance results in image classification tasks, to improve the performance of reverberant speech recognition. The DenseNet enables the deep convolutional neural network (CNN) to be effectively trained by concatenating feature maps in each convolutional layer. In addition, we extend the concept of multi-resolution CNN to multi-resolution DenseNet for robust speech recognition in reverberant environments. We evaluate the performance of reverberant speech recognition on the single-channel ASR task in reverberant voice enhancement and recognition benchmark (REVERB) challenge 2014. According to the experimental results, the DenseNet-based acoustic models show better performance than do the conventional CNN-based ones, and the multi-resolution DenseNet provides additional performance improvement.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2007.05a
• /
• pp.696-700
• /
• 2007

In case of dome-typed gymnastics training floor, since its form takes shape of the focus of sound, in the occasion when the finishing-material was used with the sandwich panel that prominent in reflexibility, and because the reverberation of sound in indoor is too loud, a smooth practice and teaching is very difficult. As this indoor gymnastics training floor, standing at its character, is required simultaneously the idea communication between the player and the coach, and the acoustic capability regarding to the clearness of music, besides the sport activity, the method to minimize the acoustic defect should be urgently contrived from the stages of design and beginning of construction. On this viewpoint, after investigation on the confidence of the surveyed value and the estimated value using acoustic simulation, and use of the changed finish-material, this thesis intends to design the dome-typed gymnastics training floor that secured the optimized acoustic condition. It is also considered that such result of the study could be utilized as the useful data that enables to improve the retrenchment effect of the construction cost as well as the acoustic performance, by means of the prediction control on the acoustic problem from the stage of design, for the occasion when the similar indoor sport gymnasium is planning to build for the near future.