• IEIE Transactions on Smart Processing and Computing
• /
• 제3권5호
• /
• pp.259-266
• /
• 2014

In this paper, we propose a novel technique for noise robust automatic speech recognition (ASR). The development of ASR techniques has made it possible to recognize isolated words with a near perfect word recognition rate. However, in a highly noisy environment, a distinct mismatch between the trained speech and the test data results in a significantly degraded word recognition rate (WRA). Unlike conventional ASR systems employing Mel-frequency cepstral coefficients (MFCCs) and a hidden Markov model (HMM), this study employ histogram of oriented gradient (HOG) features and a Support Vector Machine (SVM) to ASR tasks to overcome this problem. Our proposed ASR system is less vulnerable to external interference noise, and achieves a higher WRA compared to a conventional ASR system equipped with MFCCs and an HMM. The performance of our proposed ASR system was evaluated using a phonetically balanced word (PBW) set mixed with artificially added noise.

• 산업경영시스템학회지
• /
• 제25권1호
• /
• pp.71-77
• /
• 2002

This study deals with noise occurring in the process of human information transmission. Human process many kinds of information by receiving them from the outside or using their own thought and express the processed information to the outside by acting. In this process, Channel Noise ($C_N$) is defined as the characteristics of noise generating the incomplete result for the external stimulus. To evaluate $C_N$, keyboard-typing experiment has been performed to the 87 subjects. In this experiment, the subject consists of college students who have abundant experience in computer usage. In the experiment, there were some cases that typing order was unconsciously reversed. In this study, we deal with only these cases since these cases can be used as important data to study humans information transmission system.

• 한국멀티미디어학회논문지
• /
• 제18권7호
• /
• pp.843-852
• /
• 2015

Binarization of document images is a critical pre-processing step required for character recognition. Even though various research efforts have been devoted, the quality of binarization results largely depends on the noise amount and condition of images. We propose a new binarization method that combines Maximally Stable External Region(MSER) with down-sampling. Particularly, we propose to apply different threshold values for character regions, which turns out to be effective in reducing noise. Through a set of experiments on test images, we confirmed that the proposed method was superior to existing methods in reducing noise, while the increase of execution time is limited.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2002년도 춘계학술대회논문집
• /
• pp.579-585
• /
• 2002

This study has measured and evaluated the sound insulation of building elements of the hotel near airport. That measurement was made in both the actual site and laboratory. The differences of sound insulation performance in between granite and double glass were measured. Moreover, the difference of sound insulation was analyzed. For the measurement, this study has tried several methodologies, and analyzed the differences respectively. The results from this analysis were applied to produce a design guides for sound insulation to prevent external noise and to make more silent indoor space which satisfied the standard rate of noise.

• Progress in Superconductivity
• /
• 제4권1호
• /
• pp.68-73
• /
• 2002

We have fabricated YBCO SQUID 1st order gradiometers on $30^{\circ}$STO bicrystal substrate. The pickup coil size was 3.8mm$\times$3.8mm and baseline was 5mm. Three types of SQUID gradiometer were designed and tested for unshielded operation; solid pickup coil, pickup coil consisting of 4 parallel $ 50\mu\textrm{m}$-wide loops, and solid pickup coil with flux dam. We have investigated external magnetic field dependence of the SQUID gradiometers on the magnetic field noiseproperties. Significant increase of low frequency noise with the application of static field has been observed in the case of parallel and flux dam type pickup coil above threshold field of $1.3 \mu$T. Magnetic field noise at 1 Hz measured in the magnetically shielding room was 30, 165, 480 fTcm／sup -1／Hz／sup -1/2／ for solid type and slot type and parallel loops type, respectively.

• JSTS:Journal of Semiconductor Technology and Science
• /
• 제11권1호
• /
• pp.59-64
• /
• 2011

A low noise and low power RF front-end for 5.8 GHz DSRC (Dedicated Short Range Communication) receiver is presented. The RF front-end is composed of a single-to-differential two-stage LNA and a Gilbert down-conversion mixer. In order to remove an external balun and 5.8 GHz LC load tuning circuit, a single-to-differential LNA with capacitive cross coupled pair is proposed. The RF front-end is fabricated in a 0.13 ${\mu}m$ CMOS process and draws 7.3 mA from a 1.2 V supply voltage. It shows a voltage gain of 40 dB and a noise figure (NF) lower than 4.5 dB over the entire DSRC band.

• 대한음성학회지:말소리
• /
• 제48호
• /
• pp.141-153
• /
• 2003

Researches into robust recognition in noise environments, especially in car environments, are being carried out actively in speech community. In this paper we will report on three types of corpora that SiTEC (Speech Information TEchnology & industry promotion Center) has created for research into speech recognition in car noise environments. The first is the recordings of 900 Korean native speakers, distributed according to gender, age, and region, who uttered application words in car environments. The second is the collections of mixed noise in 3 car types by model while setting up various noise patterns which can be obtained with the car engine on or off, at different driving speed, and in different road conditions with windows open or closed. The third is the recordings of simulated speech by HATS (Head and Torso Simulator) in car environments with the internal and external noise factors added. These three types of recordings were all made through synchronized 8 channel microphones that are fixed in a car. The creation and applications of these corpora will be reported on in detail.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2014년도 춘계학술대회 논문집
• /
• pp.348-349
• /
• 2014

In general, there has been a lot of research concerned about the gear noise known to be proportional to gear transmission error for external gears likewise spur, helical gear, and hypoid gears. But, In the case of planetary gear set, gear noise study is insufficient because of the difficulty of designing, manufacturing, and understanding of the its mechanical system. This study is aimed to develop the transmission error measurement equipment for the planetary gear sets used in the automatic transmission. By comparing the results of the transmission error and noise objectively, user could select the optimized planetary gear set which has quiet noise level before manufacturing the automatic transmission.

• 한국정보통신학회논문지
• /
• 제18권4호
• /
• pp.949-956
• /
• 2014

최근, 영상정보에 관련된 하드웨어 및 소프트웨어 등의 기술이 발달함에 따라 다양한 멀티미디어 서비스에 대한 수요가 증가 되고 있다. 그러나 영상 신호의 처리, 전송, 저장하는 과정에서 여러 외부 원인에 의해 영상의 열화가 발생되며, 영상 열화의 주된 원인은 잡음에 의한 것이다. 영상은 대부분 AWGN(additive white Gaussian noise)에 의해 훼손되며, 잡음 제거를 위한 연구가 활발히 진행되고 있다. 본 논문에서는 영상에 첨가되는 AWGN의 영향을 완화하기 위하여, 공간 영역에서의 저주파 및 고주파 특성이 우수한 잡음제거 알고리즘을 제안하였다. 그리고 시뮬레이션을 통해 기존의 방법들과 결과를 비교하였으며, 성능의 평가를 위하여 PSNR(peak signal to noise ratio)을 사용하였다.

• 한국소음진동공학회논문집
• /
• 제24권7호
• /
• pp.549-554
• /
• 2014

In general, there has been a lot of research concerned about the gear noise known to be proportional to gear transmission error for external gears likewise spur, helical gear, and hypoid gears. But, In the case of planetary gear set, gear noise study is insufficient because of the difficulty of designing, manufacturing, and understanding of its mechanical system. This study is aimed to develop the transmission error measurement equipment for the planetary gear sets used in the automatic transmission. By comparing the results of the transmission error and noise objectively, user could select the optimized planetary gear set which has quiet noise level before manufacturing the automatic transmission.