• Journal of Korea Society of Industrial Information Systems
• /
• v.19 no.2
• /
• pp.73-83
• /
• 2014

Thanks to the development of diverse audio editing Technology, audio file can be easily revised. As a result, diverse social problems like forgery may be caused. Digital forensic technology is actively studied to solve these problems. In this paper, a hand-held device identification method, an area of digital forensic technology is proposed. It uses the noise features of devices caused by the design and the integrated circuit of each device but cannot be identified by the audience. Wiener filter is used to get the noise sounds of devices and their acoustic features are extracted via MIRtoolbox and then they are trained by multi-layer neural network. To evaluate the proposed method, we use 5-fold cross-validation for the recorded data collected from 6 mobile devices. The experiments show the performance 99.9%. We also perform some experiments to observe the noise features of mobile devices are still useful after the data are uploaded to UCC. The experiments show the performance of 99.8% for UCC data.

• The KIPS Transactions:PartA
• /
• v.15A no.6
• /
• pp.317-324
• /
• 2008

In this paper, we propose a system which easily controls the existing serially connected multi-channel speakers in a general personal computer by using a USB(Universal Serial Bus) interface. The personal computer as a host of the USB interface analyzes a sound source and sends audio data in a real-time fashion by the use of the isochronous transmission, one of four transmission methods provided by the USB interface. In addition, a channel is assigned by means of the bulk transmission, one of four transmission methods provided by the USB interface. Transmitted data from the USB host are sent to each speaker through compression and packet generation process. Each speaker detects corresponding digital data and regenerates audio signals through DAC(Digital-to-Analog Converter). A user can easily select a sound source file and a channel by the use of a GUI environment in a personal computer.

• The KIPS Transactions:PartD
• /
• v.9D no.2
• /
• pp.323-328
• /
• 2002

In recent years, e-Commerce is very active on the Internet, especially the World Wide Web alone: with the popularization of Internet using high-speed networks. Especially, Circulation of Multimedia Contents like MP3 data if widely being focused as one of the popular researches. However, the existing models of AOD (Audio On Demand) System lack substantial illegal copy protection or copyright protection. In this paper, we propose an AOD System that guarantees substantial illegal copy protection and copyright protection based on the PKI (Public Key Infrastructure). As transmitting MP3 data using the user's public key, the proposed .method prevents the attack of dropper during transmitting data. Also, it guarantees the right of users and distributors by prohibiting illegal users from using MP3 data.

• Communications for Statistical Applications and Methods
• /
• v.30 no.5
• /
• pp.501-516
• /
• 2023

Knowledge distillation (KD) refers to extracting knowledge from a large and complex model (teacher) and transferring it to a relatively small model (student). This can be done by training the teacher model to obtain the activation function values of the hidden or the output layers and then retraining the student model using the same training data with the obtained values. Recently, relational KD (RKD) has been proposed to extract knowledge about relative differences in training data. This method improved the performance of the student model compared to conventional KDs. In this paper, we propose a new method for RKD by introducing a new loss function for RKD. The proposed loss function is defined using the area difference between the teacher model and the student model in a specific hidden layer, and it is shown that the model can be successfully compressed, and the generalization performance of the model can be improved. We demonstrate that the accuracy of the model applying the method proposed in the study of model compression of audio data is up to 1.8% higher than that of the existing method. For the study of model generalization, we demonstrate that the model has up to 0.5% better performance in accuracy when introducing the RKD method to self-KD using image data.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2009.04a
• /
• pp.71-74
• /
• 2009

The data-flow paradigm has been employed in various application areas. It is particularly useful where large data-streams must be processed, for example in video and audio processing, or for scientific visualization. A video-based 3D reconstruction system should process multiple synchronized video streams. The system exhibits many properties that can be targeted using a data-flow approach that is naturally divided into a sequence of processing tasks. In this paper we introduce our concept to apply the data-flow approach to a multi-video 3D reconstruction system.

• Journal of The Korean Association For Science Education
• /
• v.21 no.3
• /
• pp.516-528
• /
• 2001

In the educational study, the measure of EEG(brain waves) can be useful method to study the functioning state of brain during learning behaviour. This study investigated the changes of neuronal response according to four times repetition of audio-visual learning. EEG data at the prefrontal$(Fp_{1},Fp_{2})$ were obtained from twenty subjects at the 8th grade, and analysed quantitatively using FFT(fast Fourier transform) program. The results were as follows: 1) In the first audio-visual learning, the activities of $\beta_{2}(20-30Hz)$ and $\beta_{1}(14-19Hz)$ waves increased highly, but the activities of $\theta(4-7Hz)$ and $\alpha$ (8-13Hz) waves decreased compared with the base lines. 2). According to the repetitive audio-visual learning, the activities of $\beta_{2}$ and $\beta_{1}$ waves decreased gradually after the 1st repetitive learning. And, the activity of $\beta_{2}$ wave had the higher change than that of $\beta_{1}$ wave. 3). The activity of $\alpha$ wave decreased smoothly according to the repetitive audio-visual learning, and the activity of $\theta$ wave decreased radically after twice repetitive learning. 4). $\beta$ and $\theta$ waves together showed high activities in the 2nd audio-visual learning(once repetition), and the learning achievement increased highly after the 2nd learning. 5). The right prefrontal$(Fp_{2})$ showed higher activation than the left$(Fp_{1})$ in the first audio-visual learning. However, there were not significant differences between the right and the left prefrontal EEG activities in the repetitive audio-visual learning. Based on these findings, we can conclude that the habituation of neuronal response shows up in the repetitive audio-visual learning and brain hemisphericity can be changed by learning experiences. In addition, it is suggested once repetition of audio-visual learning be effective on the improvement of the learning achievement and on the activation of the brain function.

• Journal of Advanced Marine Engineering and Technology
• /
• v.36 no.1
• /
• pp.166-171
• /
• 2012

In this paper, short-range underwater wireless communication technique using visible LEDs is proposed. As an alternative to conventional acoustic system, visible LED communications show high quality and high speed data transmission characteristics. We design a frequency modulation-based optical wireless audio transmission system. The CD4046B phase-locked loop device is applied to implement the frequency modulation and demodulation. With a transmission modulation of 100 kHz, audio signal has successfully received at a transmission distance of 30 cm.

• Journal of Broadcast Engineering
• /
• v.13 no.3
• /
• pp.415-418
• /
• 2008

Some embedded microprocessors do not have an FPU(Floating Point Unit) due to a circuit complexity and power consumption. The performance speed of MPEG-4 AAC decoder on this hardware environment would be slower than corresponding speed for playing back of the decoded results. Therefore, irritating and high-pitched noises are interleaved in the original the audio data. So, in order to play MPEG-4 AAC file on such PDA, a new algorithm that transforms floating-point arithmetic to one with integers, is needed. We have developed a transformation algorithm from floating-point operation to integer operation and implemented the PDA's AAC Player. We also show the efficiency of our proposed method with the experimental results.

• Journal of Biomedical Engineering Research
• /
• v.20 no.3
• /
• pp.275-281
• /
• 1999

Conventional PW Doppler systems suffer from the ambiguity of measured blood velocities due to the spectrum aliasing when the corresponding Doppler frequencies are greater than the Nyquist frequency. Base-line shift is a customary method for dealiasing the Doppler spectrums. I lowever, Doppler audio signals still remain unchanged even when the base-line shift method is applied. This paper de scribes an method for dealiasing both the Doppler spectra and audio signals by using sampling rate expansion, frequency shifting, and filtering poerations. For undirectional flows, the method can increase the maximum detectable Doppler frequency from the Nyquist limit of one-half of the Pulse Repetition Frequency(PRF) to the PRF. Experiments with real data have been performed to verify the proposed method.

• The Journal of the Acoustical Society of Korea
• /
• v.31 no.5
• /
• pp.317-331
• /
• 2012

In this paper, we develop a multichannel blind source separation algorithm based on a beamspace transform and the multichannel non-negative matrix factorization (NMF) method. The NMF algorithm is a famous algorithm which is used to solve the source separation problems. In this paper, we consider a beamspace-time-frequency domain data model for multichannel NMF method, and enhance the conventional method using a beamspace transform. Our decomposition algorithm is applied to audio source separation, using a dataset from the international Signal Separation Evaluation Campaign 2010 (SiSEC 2010) for evaluation.