• 대한음성학회지:말소리
• /
• 제48호
• /
• pp.81-91
• /
• 2003

In this paper, we propose a speech recognition method based on independent component analysis (ICA) and spectral enhancement techniques. While ICA tris to separate speech signal from noisy speech using multiple channels, some noise remains by its algorithmic limitations. Spectral enhancement techniques can compensate for lack of ICA's signal separation ability. From the speech recognition experiments with instantaneous and convolved mixing environments, we show that the proposed approach gives much improved recognition accuracies than conventional methods.

• 한국ITS학회 논문지
• /
• 제8권3호
• /
• pp.64-70
• /
• 2009

본 논문은 잡음에 강인한 음성인식 성능을 획득하기 위해 generalized Gamma 분포기반의 음성향상 기법을 제안한다. 우수한 음성향상을 위해서 제안된 방식에서는 generalized Gamma분포와 spectral gain floor를 이용한 음성추적 기법에 스펙트럼 최소잡음성분에 의한 희귀적인 평균 스펙트럼 값으로부터 유도되는 잡음추정을 결합하여 음질을 향상시켜 음성인식에 적용하였다. Spectral component, spectral amplitude 그리고 log spectral amplitude에 기반하여 제안된 음성향상 기법을 잡음환경에서의 음성인식에 적용하여 그 성능을 측정하였다.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1999년도 추계학술대회 논문집 학회본부 B
• /
• pp.748-750
• /
• 1999

In this study, a speech enhancement is presented based on the utilization of well-known auditory mechanism, noise masking. The speech enhancement approach adopted here is to derive an modifier that achieves audible noise suppression. This modification selectively affects the perceptually significant spectral values, and is therefore less prone to introduction of unwanted distortions than methods that affect the complete STSA and produces more enhanced results at low SNR as well as at high SNR. The speech enhancement method adopted here needs exact estimation of the minimum specteal value per critical band because it uses only the minimum spectral value per critical band. For this, the method adopted here uses the modified spectral subtraction that is more flexible than power spectral subtraction. So, the result in experiment represented better SNR than before.

• 한국음향학회지
• /
• 제36권1호
• /
• pp.64-69
• /
• 2017

본 논문에서는 PTT(Push-To-Talk) 기반의 무선 통신을 위한 진폭 및 위상 복원 기반의 단일 채널 음성 향상 방식을 제안한다. 제안한 방식은 신호의 진폭만을 대상으로 음성 향상을 진행했던 기존의 방식들과 달리, 음성 신호의 진폭과 위상을 분리하여 각각 향상시켜 다시 결합함으로써 더욱 양질의 음성을 제공한다. 본 논문에서 제안하는 방식의 성능을 평가하기 위해 동적 잡음 환경에서의 단계별 비교 실험을 실시하였으며, 실험 결과를 통해 제안한 방식이 다양한 잡음 환경에서 양질의 음성을 제공하는 것을 확인할 수 있다.

• The Journal of the Acoustical Society of Korea
• /
• 제23권4E호
• /
• pp.125-133
• /
• 2004

In this paper, we propose a critical banded wavelet packet-based spectral subtraction for speech enhancement. Critical banded wavelet packet, which reflects the human auditory system, may lead to minimization of intelligibility loss and quality improvement of the enhanced speech in the spectral domain, when combined with an appropriate spectral subtraction gain function. The proposed method shows better performance than the conventional one in comparative assessments. We also show that, for effective evaluation of enhanced speech, it is essential to consider the characteristics of speech quality measures.

• 한국자기공명학회논문지
• /
• 제15권2호
• /
• pp.157-174
• /
• 2011

Although the use of apodization functions in connection with postprocessing of a 2D NMR spectrum proves improved spectral quality, there is usually a trade-off between resolution enhancement and noise suppression due to a classical "uncertainty principle." In this study, therefore, a mathematical deconvolution technique called "Maximum Likelihood Deconvolution (MLD)" was adopted to achieve the spectral resolution and sensitivity enhancement simultaneously. The MLD technique greatly facilitates visualization and restoration of the genuine spectral information from complex 2D NMR spectra that would be problematic with the conventional apodization/FT processing. In particular, application of the MLD to the 2D-NOE spectrum would be very useful to derive the important proton connectivities, which are essential to achieve elucidating the 3D molecular structure.

• ETRI Journal
• /
• 제38권2호
• /
• pp.366-375
• /
• 2016

This paper describes an algorithm to suppress composite noise in a two-microphone speech enhancement system for robust hands-free speech communication. The proposed algorithm has four stages. The first stage estimates the power spectral density of the residual stationary noise, which is based on the detection of nonstationary signal-dominant time-frequency bins (TFBs) at the generalized sidelobe canceller output. Second, speech-dominant TFBs are identified among the previously detected nonstationary signal-dominant TFBs, and power spectral densities of speech and residual nonstationary noise are estimated. In the final stage, the bin-wise output signal-to-noise ratio is obtained with these power estimates and a Wiener post-filter is constructed to attenuate the residual noise. Compared to the conventional beamforming and post-filter algorithms, the proposed speech enhancement algorithm shows significant performance improvement in terms of perceptual evaluation of speech quality.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2003년도 학술회의 논문집 정보 및 제어부문 B
• /
• pp.511-514
• /
• 2003

The human being uses speech signals to exchange information. When background noise is present, speech recognizers experience performance degradations. Speech recognition through speech enhancement in the noisy environment was studied. Histogram method as a reliable noise estimation approach for spectral subtraction was introduced using MFCC method. The experiment results show the effectiveness of the proposed algorithm.

• 한국정보통신학회논문지
• /
• 제13권10호
• /
• pp.2211-2217
• /
• 2009

본 논문에서는 잔향이 존재하는 환경에서 단일 마이크로폰을 사용한 음성 개선 방법을 제시한다. 스펙트럼 차감법(Spectral Subtraction)은 스펙트럼 상에서 잔향성분 및 잡음을 제거 할 수 있는 효과적인 방법이다. 스펙트럼 차감법은 음성과 비음성 구간의 정확한 구분을 필요로 하며 성능을 향상시키기 위해 본 논문에서는 엔트로피(Entropy) 기반의 음성 구간 검출법을 적용하였다. 제시된 방법을 기존의 에너지 검출 기반의 음성 검출법을 적용한 스펙트럼 차감법과 비교하여 성능 평가를 수행하였다. SNR 및 잔향시간에 따른 잔향 제거비율을 평가지표로 사용하였으며, 시뮬레이션 결과 기존의 스펙트럼 차감법과 비교하여 제시된 방법이 우수한 성능을 보였다.

• 대한기계학회:학술대회논문집
• /
• 대한기계학회 2004년도 추계학술대회
• /
• pp.607-612
• /
• 2004

Spectral resolution enhancement method of Acousto-Optic Tunable Filter (AOTF) using incident light angle variation is described. AOTF is a small, mechanically rigid, high speed and spectral resolution light tunable filter. The basic theory of AOTF and its experimental verification is described. AOTF can generate two opposite polarized light simultaneously which wavelength can be changed by incident angle variation. We focused on the common region of two filtered light at the specific incident angle. This region can be used to enhance the spectral resolution of AOTF.