• 한국지능시스템학회논문지
• /
• 제20권5호
• /
• pp.683-687
• /
• 2010

본 논문에서는 인간의 감정 변화에 강인한 음성 인식 시스템을 구현하기 위하여 음성 변환 방법 중의 한가지인 주파수 와핑 방법을 사용한 연구를 수행하였다. 이러한 목표를 위하여 다양한 감정이 포함된 음성 데이터베이스를 사용하여 감정의 변화에 따라 음성의 스펙트럼이 변화한다는 것과 이러한 변화는 음성 인식 시스템의 성능을 저하시키는 원인 중의 하나임을 관찰하였다. 본 논문에서는 이러한 음성의 변화를 감소시키는 방법으로 주파수 와핑을 학습 과정에 사용하는 방법을 제안하여 감정 변화에 강인한 음성 인식 시스템을 구현하였고 성도 길이 정규화 방법을 사용한 방법과 성능을 비교하였다. HMM을 사용한 단독음 인식 실험에서 제안된 학습 방법은 사용하면 감정이 포함된 데이터에 대한 인식 오차가 기존 방법보다 감소되었다.

• 한국음향학회지
• /
• 제29권5호
• /
• pp.339-346
• /
• 2010

음향 간섭제거기는 머리전달함수 (head-related impulse response; HRIR)의 길이에 큰 영향을 받게 되어, 일반적으로 큰 차수의 필터를 필요로 한다. 간섭제거필터의 길이를 줄이기 위한 방법으로 주파수 워핑, 공통 극점과 영점 (common acoustical pole and zero; CAPZ) 모델링 등의 방법들이 제안되었는데, 본 논문에서는 이 두 가지 방법을 결합한 방법을 제안한다. 이를 위해, 주파수 워핑 영역에서 공통 극점과 영점 모델링을 통해 필터를 설계하며, 디워핑 과정을 통해 종래의 선형 영역에서 안정된 필터를 구현한다. 제안된 방법은 주파수 워핑을 통한 간섭제거 성능 향상과 공통 극점 모델링을 통한 필터 계수 감소를 함께 제공할 수 있다. 이러한 성능을 검증하기 위해 다양한 컴퓨터 모의 실험을 진행하였다.

• 한국통신학회논문지
• /
• 제24권12B호
• /
• pp.2407-2414
• /
• 1999

본 논문에서는 WDCT(Warped Discrete Cosine Transform)의 개념에 대해서 소개하고 이의 응용분야로서 WDCT를 이용한 영상 압축 알고리듬을 제시한다. WDCT는 기존의 일반적인 DCT와 주파수 특성이 하나의 파라미터로 조절되는 IIR(infinte impulse response) 전대역 통과 필터(all-pass filter)를 직렬로 연결한 변환이다. 제시된 영상 압축 알고리듬에서는 필터의파라미터가 미리 정의된 범위 내에서 조절되도록 한다. 각 영상의 블록에 대해서 주어진 범위 내에서 가장 좋은 파라미터가 선정되면 이를 이용한 WDCT의 결과와 이 파라미터를 디코더로 전송한다. 본 논문에서는 IIR 전대역 통과 필터링 과정을 하나의 행렬로 대체하거나 DCT를 필터뱅크로 보아 IIR 필터와 DCT의 결합을 일반적인 DCT와 마찬가지로 하나의 행렬로 표현하였다. 따라서 주어진 파라미터에 따라 각각 다른 새로운 WDCT 행렬을 정의할 수 있으므로 WDCT의 결과는 행렬과 벡터의 곱으로 얻어진다. WDCT를 이용한 영상 압축의 결과는 높은 비트율과 고주파 성분이 많은 영상에 대하여 DCT의 성능보다 우수함을 알 수 있었다.

• 한국지능시스템학회논문지
• /
• 제11권6호
• /
• pp.563-567
• /
• 2001

We generalize the widebane P0-weyl symbol (P0WS) and the widebane spreading function (WSF) using the generalized warping function . The new generalized P0WS and WSF are useful for analyzing system and communication channels producing generalized time shifts. We also investigated the relationship between the affine Wey1 symbol(AWS) and the P0WS. By using specific warping functions, we derive new P0WS and WSF as analysis tools for systems and communication channels with non-linear group delary characteristics. The new P0WS preserves specific types of changes imposed on random processes. The new WSF provides a new interpretation of output of system and communication channel as weighted superpositions of non-linear time shifts on the input. It is compared to the conventional method obtaining output of system and communication channel as a convention integration of the input with the impulse response of the system and the communication channel. The convolution integration can be interpreted as weighted superpositions of liner time shifts on the input where the weight is the impulse response of the system and the communication channel. Application examples in analysis and detection demonstrate the advantages of our new results.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제7권2호
• /
• pp.102-108
• /
• 2007

The issues of preventive and condition-based maintenance, online monitoring, system fault detection, diagnosis, and prognosis are of increasing importance. This study introduces a technique to detect and identify faults in induction motors. Stator currents were measured and stored by time domain. The time domain is not suitable for representing current signals, so wavelet transform is used to convert the signal; onto frequency domain. The raw signals can not show the significant feature, therefore difference values are applied. The difference values were transformed by wavelet transform and the features are extracted from the transformed signals. The dynamic time warping method was used to identify the four fault types. This study describes the results of detecting fault using wavelet analysis.

• Journal of Advanced Marine Engineering and Technology
• /
• 제31권1호
• /
• pp.95-102
• /
• 2007

The issues of preventive and condition-based maintenance, online monitoring, system fault detection, diagnosis and prognosis are of increasing importance. This study introduces a technique to detect and identify faults in induction motors. Stator currents were measured and stored by time domain. The time domain is not suitable for representing current signals, so wavelet transform is used to convert the signals onto frequency domain. The raw signals can not show the significant feature, therefore difference values between the signal of the health conditions and that of the fault conditions are applied. The difference values were transformed by wavelet transform and the features are extracted from the transformed signals. The dynamic time warping method was used to identify the fault type. This study describes the results of detecting fault using wavelet analysis.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2011년도 추계학술대회 논문집
• /
• pp.130-135
• /
• 2011

A study of the bending-extension-transverse shear coupled random response of the composite beams with thin-walled open sections subjected to various types of concentrated and distributed random excitations is dealt with in this paper. First of all, equations of motion of thin-walled composite H-type cross-section beams incorporating a number of nonclassical effects of transverse shear and primary and secondary warping, and anisotropy of constituent materials are derived. On the basis of derived equations of motion, analytical expressions for the displacement response of the composite beams are derived by using normal mode method combined with frequency response function method.

• 소음진동
• /
• 제9권6호
• /
• pp.1193-1199
• /
• 1999

This paper deals with the free vibrations of circular curved beams with thin-walled cross-section. The differential equation for the coupled flexural-torsional vibrations of such beams with warping is solved numerically to obtain natural frequencies and mode shapes. The Runge-Kutta and determinant search methods, respectively, are used to solve the governing differential equation and to compute the eigenvalues. The lowest three natural frequencies and corresponding mode shapes are calculated for the thin-walled horizontally curved beams with hinged-hinged, hinged-clamped, and clamped-clamped end constraints. A wide range of opening angle of beam, warping parameter, and two different values of slenderness ratios are considered. Numerical results are compared with existing exact and numerical solutions by other methods.

• 말소리와 음성과학
• /
• 제7권1호
• /
• pp.3-10
• /
• 2015

In this paper, different frequency scales in cepstral feature extraction are evaluated for the text-independent speaker recognition. To this end, mel-frequency cepstral coefficients (MFCCs), linear frequency cepstral coefficients (LFCCs), and bilinear warped frequency cepstral coefficients (BWFCCs) are applied to the speaker recognition experiment. In addition, the spectro-temporal features extracted by the cepstral-time matrix (CTM) are examined as an alternative to the delta and delta-delta features. Experiments on the NIST speaker recognition evaluation (SRE) 2004 task are carried out using the Gaussian mixture model-universal background model (GMM-UBM) method and the joint factor analysis (JFA) method, both based on the ALIZE 3.0 toolkit. Experimental results using both the methods show that BWFCC with appropriate warping factor yields better performance than MFCC and LFCC. It is also shown that the feature set including the spectro-temporal information based on the CTM outperforms the conventional feature set including the delta and delta-delta features.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2008년도 학술대회 논문집 정보 및 제어부문
• /
• pp.290-291
• /
• 2008

The most popular speech feature used in speech recognition today is the Mel-Frequency Cepstral Coefficients (MFCC) algorithm, which could reflect the perception characteristics of the human ear more accurately than other parameters. This paper adopts MFCC and its first order difference, which could reflect the dynamic character of speech signal, as synthetical parametric representation. Furthermore, we quote Dynamic Time Warping (DTW) algorithm to search match paths in the pattern recognition process. We use the software "GoldWave" to record English digitals in the lab environments and the simulation results indicate the algorithm has higher recognition accuracy than others using LPCC, etc. as character parameters in the experiment for Digital Isolated Word Recognition (DIWR) system.