• Proceedings of the Acoustical Society of Korea Conference
• /
• 1993.06a
• /
• pp.128-131
• /
• 1993

정적지도 화자적응기법에서 스펙트럼 거리에 의존하는 비선형적인 스펙트럼사상법을 이용하여 잡음환경에서의 음성인식방법에 관하여 연구한 결과, Top2에서 인식율의 향상을 얻어 그 유효성을 확인하였다. 본 연구에서는 스펙트럼 거리에 의존하지 않는 선형 스펙트럼 사상법을 제시하고 그에 의한 잡음환경의 음성인식결과를 비선형적인 스펙트럼 사상법에 의한 결과와 비교하였다. 그 결과, 인식율이 개선되었을 뿐만 아니라, Top1에서도 인식율이 향상되어 선형 스펙트럼사상법이 잡음환경음성인식방법으로 효과적인 방법임을 확인하였다.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.26 no.10B
• /
• pp.1360-1366
• /
• 2001

이 논문에서는, 잡음환경에서의 음성 향상을 위해서 일반적인 주파수 차감법에 기반한 새로운 형태의 방법을 제안한다. 기존의 방법들이 각각의 주파수 성분에 대해 잡음 및 음성스펙트럼을 추정하는데 비해, 본 논문에서는 주파수 대역을 여러 개의 서브밴드로 대역을 나누어 각각의 서브밴드에 대해서 잡음 및 음성의 스펙트럼을 추정한다. 본 논문에서는 잡음 스펙트럼을 추정하기 위하여 최소추적(Minima Tracking) 방법을 선택하였고, 필터링 방법으로는 스펙트럼 차감법에 기반한 Mel-Scaled 필터뱅크를 이용한 새로운 방법을 제안하였다. 모의실험결과, 기존의 방법들에 비해 음성구간에서의 SNR의 향상정도는 입력 SNR이 -10∼4dB의 범위에서 향상된 결과를 얻었다. 또한 전 구간에 대해서도 다른 알고리즘들 보다 향상된 결과를 얻었다.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1998.06e
• /
• pp.343-346
• /
• 1998

잡음환경에서의 음성인식 성능향상을 위하여 본 논문에서는 스펙트럼 차감법 이후에 남아 있는 잔여 잡음으로 인한 mismatch를 극복하는 수단으로 기존의 스펙트럼 차감법에서의 flooring factor를 사용하는 대신에 target 잡음레벨을 이용하여 잡음 마스킹을 적용하는 스펙트럼 차감법과 잡음 마스킹의 hybrid 방식을 사용한다. 이 방법은 낮은 SNR에서 개선되지 않는 기존의 잡음 마스킹이 가지는 약점을 극복하고 동시에 스펙트럼 차감버에서의 잔여 잡음 문제를 완화시킬 수 있었다. 특히 시간/주파수 영역 smoothing을 적용함으로써 스펙트럼 차감법과 잡음 마스킹의 hybrid 방식의 적용 이후에도 여전히 남아 있는 일부 잡음을 추가적으로 감소시켰으며, 더욱 향상된 인식성능을 얻을 수 있었다.

• Journal of Satellite, Information and Communications
• /
• v.12 no.2
• /
• pp.46-50
• /
• 2017

In this paper, we propose an artificial intelligence based spectrum allocation scheme for power line communication system. The frequency band of the transmitted signal can be adjusted through the spectrum allocation technique, thereby avoiding interference. This improves the performance of the transmission signal and the spectral efficiency. Through the simulation results, we show that the proposed spectrum allocation technique improves the spectral efficiency and improve the communication performance.

• The Journal of the Acoustical Society of Korea
• /
• v.41 no.1
• /
• pp.38-44
• /
• 2022

Speech enhancement is performed to improve intelligibility and quality of the noise-corrupted speech. In this paper, speech enhancement performance was compared using different loss functions in time and frequency domains. This study proposes a combination of loss functions to utilize advantage of each domain by considering both the details of spectrum and the speech waveform. In our study, Scale Invariant-Source to Noise Ratio (SI-SNR) is used for the time domain loss function, and Mean Squared Error (MSE) is used for the frequency domain, which is calculated over the complex-valued spectrum and magnitude spectrum. The phase loss is obtained using the sin function. Speech enhancement result is evaluated using Source-to-Distortion Ratio (SDR), Perceptual Evaluation of Speech Quality (PESQ), and Short-Time Objective Intelligibility (STOI). In order to confirm the result of speech enhancement, resulting spectrograms are also compared. The experimental results over the TIMIT database show the highest performance when using combination of SI-SNR and magnitude loss functions.

• Journal of Advanced Navigation Technology
• /
• v.24 no.2
• /
• pp.121-127
• /
• 2020

In this paper, we propose a dynamic spectrum hole grouping method that changes the grouping range of spectrum hole according to the resources amount required by secondary users in wideband spectrum cognitive wireless networks, and then the proposed method is applied to channel allocation for the secondary user service. The proposed method can improve waste of resources in the existing static spectrum hole grouping in virtue of grouping dynamically as much the predicted spectrum holes resources as secondary users require. Simulation results show that channel allocation method with the proposed dynamic grouping outperforms that with the static grouping method in resources utilization under acceptable secondary user service performance.

• The Journal of the Acoustical Society of Korea
• /
• v.42 no.6
• /
• pp.544-551
• /
• 2023

Speech enhancement used to improve the perceptual quality and intelligibility of noise speech has been studied as a method using a complex-valued spectrum that can improve both magnitude and phase in a method using a magnitude spectrum. In this paper, a study was conducted on how to apply attention mechanism to complex-valued spectrum-based speech enhancement systems to further improve the intelligibility and quality of noise speech. The attention is performed based on additive attention and allows the attention weight to be calculated in consideration of the complex-valued spectrum. In addition, the global average pooling was used to consider the importance of the feature map. Complex-valued spectrum-based speech enhancement was performed based on the Deep Complex U-Net (DCUNET) model, and additive attention was conducted based on the proposed method in the Attention U-Net model. The results of the experiments on noise speech in a living room environment showed that the proposed method is improved performance over the baseline model according to evaluation metrics such as Source to Distortion Ratio (SDR), Perceptual Evaluation of Speech Quality (PESQ), and Short Time Object Intelligence (STOI), and consistently improved performance across various background noise environments and low Signal-to-Noise Ratio (SNR) conditions. Through this, the proposed speech enhancement system demonstrated its effectiveness in improving the intelligibility and quality of noisy speech.

• The Journal of the Acoustical Society of Korea
• /
• v.29 no.7
• /
• pp.468-474
• /
• 2010

The major factor that disturbs practical use of speech recognition is distortion by the ambient and channel noises. Generally, the ambient noise drops the performance and restricts places to use. DSR (Distributed Speech Recognition) based speech recognition also has this problem. Various noise cancelling algorithms are applied to solve this problem, but loss of spectrum and remaining noise by incorrect noise estimation at low SNR environments cause drop of recognition rate. This paper proposes methods for speech enhancement. This method uses MMSE-STSA for noise cancelling and ideal binary mask to compensate damaged spectrum. According to experiments at noisy environment (SNR 15 dB ~ 0 dB), the proposed methods showed better spectral results and recognition performance.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1998.06e
• /
• pp.391-394
• /
• 1998

본 논문에서는 극심한 가산 잡음에 의해 손상된 음성 신호를 스펙트럼 차감법으로 개선할 때, 잡음 스펙트럼 추정을 위한 무음 구간 추정 방법을 제안한다. 스펙트럼 차감법은 잡음을 효과적으로 제거한다고 알려져 있으나, SNR 0 dB 이하의 잡음 환경에서는 무음 구간의 검출이 힘들어 잡음 스펙트럼 추정치의 정확도가 저하된다. 일반화 스펙트럼 차감법의 과차감(oversubtraction)과 잡음 스펙트럼 추정을 반복하여 얻은 무음 구간은 SNR -10 dB~ 0 dB의 낮은 SNR에서도 비교적 정확하며, 프레임 에너지를 이용한 무음 검출 방법에 비해 향상된 성능을 보였다.

• Journal of Advanced Navigation Technology
• /
• v.23 no.4
• /
• pp.322-327
• /
• 2019

In this paper, we propose first a scheme of grouping spectrum holes that are created in the multiple channels of primary users, and then by using the scheme we enhance quality of service (QoS) of wideband cognitive radio users in cellular cognitive radio networks. In our scheme, spectrum holes created in each primary channel are predicted by Wiener prediction process, and then the predicted spectrum holes happened in the same time are grouped into a group. The wideband cognitive radio users explore the group of spectrum holes to improve their QoS. Simulation results show that their handoff calls dropping rate and initial calls blocking rate are significantly reduced in our scheme, compared to those in the single primary channel.