• Proceedings of the Korea Information Processing Society Conference
• /
• 2014.11a
• /
• pp.693-694
• /
• 2014

This short paper briefly describes the proposed music recommendation method that provides suitable music pieces to a listener depending on both listeners' ratings and content of music pieces. The proposed method consists of two methods. First, listeners' ratings prediction method is a combination the traditional user-based and item-based collaborative filtering methods. Second, genre classification method is a combination of feature extraction and classification procedures. The feature extraction step obtains audio signal information and stores it in data structure, while the second one classifies the music pieces into various genres using decision tree algorithm.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2018.10a
• /
• pp.746-748
• /
• 2018

전 세계적으로 인플루엔자에 의해 매년 29~64만의 사망자가 발생하며 사회, 경제적 피해를 일으키고 있다. 기침에 의해 생성된 비말은 인플루엔자의 주요 전파 방법으로, 기침 감지 기술을 통해 확산 방지가 가능하다. 이전의 기침 감지에 대한 연구는 기침 소리와 전통적인 기계학습기법을 사용하였다. 본 논문은 기침 소리와 더불어 기침 시 발생하는 신체의 움직임 정보를 동시에 학습하는 멀티모달 딥러닝 기반의 기침 감지 모델을 제안한다. 도출된 모델과 기존의 모델과의 성능 비교를 통해 제안한 모델이 이전의 기침 감지 모델보다 정확한 기침 인식이 가능함을 보였다. 본 논문이 제안하는 모델은 스마트 워치와 같은 웨어러블 기기에 적용되면 인플루엔자의 확산 방지에 크게 기여할 수 있을 것이다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2000.04a
• /
• pp.373-378
• /
• 2000

ARS(Audio Response System)를 구현하기 위하여, 응용 분야를 분석하여 필요한 자료구조를 설계 및 처리 과정을 설계한 후, ARS 처리용 하드웨어 생산 업체에서 제공하는 원시 라이브러리와 C언어를 사용하여 구현하는 것이 일반적이다. 본 논문에서는 ARS의 처리 과정을 분석하여 동작을 제어하는 부분과 동작을 표현하는 부분으로 분리한 ARS 구현 모델을 제시하였다. 응용분야와 무관한 동작제어 부분은 대기상태, 처리상태, 종료상태로 구성되는 유한상태 기계 모델을 제시하였고, 응용분야에 따라 결정되는 동작표현에 필요한 정보를 체계적으로 구성한 자료구조를 제시하였다. 본 논문에서 제시하는 모델에 의하여 동작표현만 제공함으로써 ARS를 구현할 수 있다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.168-169
• /
• 2023

머신 러닝의 학습을 위한 데이터는 개인정보가 포함된 데이터인 경우가 존재한다. 특히 음성인식 모델을 학습시키기 위해서 사용자의 음성 데이터가 필요하며, 이는 개인의 민감한 정보가 포함될 수 있다. 인공지능 학습을 위해 수집한 음성 데이터에 대한 정보보호 침해 공격이 발생할 수 있고, 해당 데이터에 대한 보호 조치가 필요하다. 본 연구는 음성 데이터를 안전하게 관리하기 위해 분할학습을 이용한 음성 데이터 학습 모델을 제안한다.

• Proceedings of the IEEK Conference
• /
• 2006.06a
• /
• pp.257-258
• /
• 2006

In this paper, we address several issues in the real time implementation of MPEG-1 Layer II decoder on a fixed-point digital signal processor (DSP), especially TMS320C6416. There is a trade-off between processing speed and the size of program/data memory for the optimal implementation. In a view of the speed optimization, we first convert the floating point operations into fixed point ones with little degradation in audio quality, and then the look-up tables used for the inverse quantization of the audio codec are forced to be located into the internal memory of the DSP. And then, window functions and filter coefficients in the decoder are precalculated and stored as constant, which makes the decoder faster even larger memory size is required. It is shown from the real-time experiments that the fixed-point implementation enables us to make the decoder with a sampling rate of 48 kHz operate with 3 times faster than real-time on TMS320C6416 at a clock rate of 600 MHz.

• The Journal of the Acoustical Society of Korea
• /
• v.26 no.5
• /
• pp.188-193
• /
• 2007

In general, analog amplifiers for 5.1 or more channels have been used to configure home theater systems. In order to make high-performance systems, it is desirable to process audio signals in digital techniques in consideration of output and efficiency of speakers. Especially we need 7.1-channel system to separate audio signals efficiently. In this paper we implemented the architecture of S/PDIF 7.1-channel digital amplifier for home theater systems. The amplifier shows good performance with less loss of original sounds because of both strong characteristics against noises and direct processing of input data.

• Journal of the Institute of Convergence Signal Processing
• /
• v.19 no.2
• /
• pp.47-53
• /
• 2018

This paper proposes a method to express an MP3 audio file by a series of color QR codes which can be printed on the paper. Moreover, the method can compress the data considerably. Firstly, an MP3 file is divided into many small files which have maximum capacity of binary file of a QR code. Secondly, the multiple files are converted to multiple black-and-white QR codes. Lastly, every three QR codes are combined into color QR codes. When combining, each of three black-and-white QR codes are regarded as red, green, blue components respectively. In this method, the areas of a color QR code where two QR codes are overlapped are expressed by the colors Cyan, Magenta and Yellow. And the areas where three components are overlapped are expressed by white color. Contrarily, the areas that no components are overlapped are expressed by white color. Experimentation result shows that an MP3 file with 8.5MB the original MP3 files are compressed with the compression rate around 15.7. This method has the advantage that can be used in the environments that the internet access is impossible.

• The Journal of the Acoustical Society of Korea
• /
• v.28 no.1
• /
• pp.1-9
• /
• 2009

This paper deals with the realization of multi-channel live karaoke. In this study, 6-channel MP3 decoding and tempo/key scaling was operated in real time by using the TMS320C6713 DSP, which is 32 bit floating-point DSP made by TI Co. The 6 channel consists of front L/R instrument, rear L/R instrument, melody, and woofer. In case of the 4 channel, rear L/R instrument can be replaced with drum L/R channel. And the final output data is generated as adjusted to a 5.1 channel speaker. The SOLA algorithm was applied for tempo scaling, and key scaling was done with interpolation and decimation in the time domain. Drum channel was excluded in key scaling by separating instruments into drums and non-drums, and in processing SOLA, high-quality tempo scaling was made possible by differentiating SOLA frame size, which was optimized for real-time process. The use of 6 channels allows the composition of various channels, and the multi-channel audio system of this study can be effectively applied at any place where live music is needed.

• Journal of Information Processing Systems
• /
• v.13 no.3
• /
• pp.599-617
• /
• 2017

Due to the rapid growth and expansion of the Internet, the digital multimedia such as image, audio and video are available for everyone. Anyone can make unauthorized copying for any digital product. Accordingly, the owner of these products cannot protect his ownership. Unfortunately, this situation will restrict any improvement which can be done on the digital media production in the future. Some procedures have been proposed to protect these products such as cryptography and watermarking techniques. Watermarking means embedding a message such as text, the image is called watermark, yet, in a host such as a text, an image, an audio, or a video, it is called a cover. Watermarking can provide and ensure security, data authentication and copyright protection for the digital media. In this paper, a new watermarking method of still image is proposed for the purpose of copyright protection. The procedure of embedding watermark is done in a transform domain. The discrete cosine transform (DCT) is exploited in the proposed method, where the watermark is embedded in the selected coefficients according to several criteria. With this procedure, the deterioration on the image is minimized to achieve high invisibility. Unlike the traditional techniques, in this paper, a new method is suggested for selecting the best blocks of DCT coefficients. After selecting the best DCT coefficients blocks, the best coefficients in the selected blocks are selected as a host in which the watermark bit is embedded. The coefficients selection is done depending on a weighting function method, where this function exploits the values and locations of the selected coefficients for choosing them. The experimental results proved that the proposed method has produced good imperceptibility and robustness for different types of attacks.

• The Journal of the Acoustical Society of Korea
• /
• v.32 no.6
• /
• pp.509-517
• /
• 2013

In this paper, we researched a method that makes a good acoustic-speech system using a digital signal processing technique with dynamic microphone as a transducer. Good acoustic-speech system should deliver the original sound input to electric signal without distortion. By measuring the frequency response of the microphone, adjustment factors are obtained by comparing measured data and standard frequency response of microphone for each frequency band. The final sound levels are obtained using the developed adjustment factors of frequency responses from the microphone and speaker to match the original sound levels using the digital signal processing technique. Then, we minimize the changes in the frequency response and level due to the variation of the distance from source to microphone, where the frequency responses were measured according to the distance changes.