http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Noise Robust Automatic Speech Recognition Scheme with Histogram of Oriented Gradient Features
Park, Taejin,Beack, SeungKwan,Lee, Taejin The Institute of Electronics and Information Engin 2014 IEIE Transactions on Smart Processing & Computing Vol.3 No.5
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.
A Spatial Audio System Using Multiple Microphones on a Rigid Sphere
Taejin Lee,강경옥,김진웅,정대권,Daeyoung Jang,Hareo Hamada 한국전자통신연구원 2005 ETRI Journal Vol.27 No.2
The main purpose of a spatial audio system is to give a listener the same impression as if he/she were present in a recorded environment. A dummy head microphone is generally used for such purposes. Because of its humanlike shape, we can obtain good spatial sound images. However, its shape is a restriction on its public use and it is difficult to convert a 2-channel recording into multichannel signals for an efficient rendering over a multispeaker arrangement. In order to solve the problems mentioned above, a spatial audio system is proposed that uses multiple microphones on a rigid sphere. The system has five microphones placed on special points of the rigid sphere, and it generates audio signals for headphone, stereo, stereo dipole, 4-channel, and 5-channel reproduction environments. Subjective localization experiments show that front/back confusion, which is a common limitation of spatial audio systems using the dummy head microphone, can be reduced dramatically in 4-channel and 5-channel reproduction environments and can be reduced slightly in a headphone reproduction
Effect of Surface Oxidation Layer on Tensile Strength of Cu-Ni Alloy in Friction Stir Welding
Taejin Yoon,Sangwon Park,Sungwook Chung,Joongsuk Noh,Kwangho Kim,Chungyun Kang 대한금속·재료학회 2016 METALS AND MATERIALS International Vol.22 No.3
Friction stir welding (FSW) of thick Cu-Ni plate was successfully completed. The fracture position after tensile testing was located at the weld nugget zone (WNZ), where surface oxidation occurred. The oxidation morphologies on the surface of the base metal were analyzed by SEM, EPMA and XRD, with the oxide layer being obtained by simple and useful way to analyze the oxide products, namely, collecting oxide powders after immersing of the oxidized specimen into HNO3 solution. The results highlighted that an oxide layer of 30 μm thickness consists of a mixture of two phases, Cu2O and NiO, on the surface of the base metal. After FSW, the thickness of the oxide layer on the surface was decreased to approximately 5 μm, and broken oxide particles, which is NiO, penetrated into the WNZ by the rotating tool. NiO was preferentially formed at the surface after FSW because it has a lower Gibbs free energy value at 950 °C, which is the peak temperature measured during FSW. Oxide layer of Cu-Ni plate was clearly only removed by mechanical method grinding with 1200-grit SiC paper. The removal of oxide layer results in improved mechanical strength.
Boundary String Field Theory and an Open String One-Loop
Taejin Lee,K. S. Viswanathan,Yi Yang 한국물리학회 2003 THE JOURNAL OF THE KOREAN PHYSICAL SOCIETY Vol.42 No.1
We discuss the open string one-loop partition function in the tachyon condensation background of an unstable D-brane system. We evaluate the partition function by using the boundary-state formulation and find that it is in complete agreement with the result obtained in the boundary string eld theory. This suggests that the open string higher loop diagrams may be produced consistently by using a closed string field theory, where the D-brane plays the role of a source for the closed string field.
Optimized Implementation of Interpolation Filters for HEVC Encoder
Taejin Hwang,안용조,유지우,심동규 대한전자공학회 2013 전자공학회논문지 Vol.50 No.11
In this paper, a fast algorithm of discrete cosine transform-based interpolation filter (DCT-IF) for HEVC (high efficiency video coding) encoder is proposed. DCT-IF filter accounts for around 30% of encoder complexity, according to the computational complexity analysis with the HEVC reference software. In this work, the proposed DCT-IF is optimized by applying frame-level interpolation, SIMD optimization, and task-level parallelization via OpenMP on a developed C-based HEVC encoder. Performance analysis is conducted by measuring speed-up factor of the proposed optimization technique on the developed encoder. The results show that speed-up factors by frame-level interpolation, SIMD, and OpenMP are approximately 38-46, 3.6-4.4, and 3.0-3.7, respectively. In the end, we achieved the speed-up factor of 498.4 with the proposed fast algorithm.