• ETRI Journal
• /
• 제38권6호
• /
• pp.1190-1196
• /
• 2016

We propose a new bandpass filter (BPF)-based online channel normalization method to dynamically suppress channel distortion when the speech and channel noise components are unknown. In this method, an adaptive modulation frequency filter is used to perform channel normalization, whereas conventional modulation filtering methods apply the same filter form to each utterance. In this paper, we only normalize the two mel frequency cepstral coefficients (C0 and C1) with large dynamic ranges; the computational complexity is thus decreased, and channel normalization accuracy is improved. Additionally, to update the filter weights dynamically, we normalize the learning rates using the dimensional power of each frame. Our speech recognition experiments using the proposed BPF-based blind channel normalization method show that this approach effectively removes channel distortion and results in only a minor decline in accuracy when online channel normalization processing is used instead of batch processing

• 한국통신학회논문지
• /
• 제39A권3호
• /
• pp.155-157
• /
• 2014

In this paper, a novel codebook generation strategy is proposed. With the given codebooks, two codeword selection procedures are proposed and analyzed for generating the quantized multiple-input multiple-output (MIMO) channel state information (CSI). Furthermore, three different quantization and normalization strategies are analyzed. The simulation results suggest that the proposed 'quantized channel generation method 2' is the best strategy to reduce the quantization and normalization errors to generate the final quantized MIMO CSI.

• 한국음향학회지
• /
• 제35권4호
• /
• pp.280-287
• /
• 2016

본 논문에서는 다채널 능동 소음 제어를 위한 적응 필터에 적용할 수 있는 정규화된 FxLMS(Filtered-x Least Mean Square) 알고리즘을 제안하였다. 단일 채널 능동 소음 제어를 위한 FxLMS 알고리즘의 경우는 기존의 NLMS(Normalized Least Mean Square) 알고리즘과 같은 방식으로 정규화할 수 있는 반면, 다채널 능동 소음 제어의 경우에는 단일 채널 방식의 정규화 알고리즘을 그대로 적용할 수 없다. 먼저, 최소 교란 원리에 근거한 일반화된 정규화 알고리즘을 이용하여, 역행렬 연산을 피하기 위하여 대각 성분만을 고려한 정규화 알고리즘을 제안하였다. 컴퓨터 모의 실험을 통하여 제안된 알고리즘을 정규화되지 않은 기존의 알고리즘들과 비교하였다. 제안된 알고리즘이 정규화되지 않은 기존의 알고리즘에 비하여 비정상 환경에서 우수한 성능을 가진다는 것을 보였다.

• ETRI Journal
• /
• 제29권3호
• /
• pp.300-304
• /
• 2007

This paper proposes a new data-driven method for high-pass approaches, which suppresses slow-varying noise components. Conventional high-pass approaches are based on the idea of decorrelating the feature vector sequence, and are trying for adaptability to various conditions. The proposed method is based on temporal local decorrelation using the information-maximization theory for each utterance. This is performed on an utterance-by-utterance basis, which provides an adaptive channel normalization filter for each condition. The performance of the proposed method is evaluated by isolated-word recognition experiments with channel distortion. Experimental results show that the proposed method yields outstanding improvement for channel-distorted speech recognition.

• 대한음성학회지:말소리
• /
• 제44호
• /
• pp.73-82
• /
• 2002

Channel distortion degrades the performance of speech recognizer in telephone environment. It mainly results from the bandwidth limitation and variation of transmission channel. Variation of channel characteristics is usually represented as baseline shift in the cepstrum domain. Thus undesirable effect of the channel variation can be removed by subtracting the mean from the cepstrum. In this paper, to improve the recognition performance of Korea connected digit telephone speech, channel compensation methods such as CMN (Cepstral Mean Normalization), RTCN (Real Time Cepatral Normalization), MCMN (Modified CMN) and MRTCN (Modified RTCN) are applied to the static MFCC. Both MCMN and MRTCN are obtained from the CMN and RTCN, respectively, using variance normalization in the cepstrum domain. Using HTK v3.1 system, recognition experiments are performed for Korean connected digit telephone speech database released by SITEC (Speech Information Technology & Industry Promotion Center). Experiments have shown that MRTCN gives the best result with recognition rate of 90.11% for connected digit. This corresponds to the performance improvement over MFCC alone by 1.72%, i.e, error reduction rate of 14.82%.

• IEIE Transactions on Smart Processing and Computing
• /
• 제1권3호
• /
• pp.143-151
• /
• 2012

A new data-driven method for the design of a blind modulation frequency filter that suppresses the slow-varying noise components is proposed. The proposed method is based on the temporal local decorrelation of the feature vector sequence, and is done on an utterance-by-utterance basis. Although the conventional modulation frequency filtering approaches the same form regardless of the task and environment conditions, the proposed method can provide an adaptive modulation frequency filter that outperforms conventional methods for each utterance. In addition, the method ultimately performs channel normalization in a feature domain with applications to log-spectral parameters. The performance was evaluated by speaker-independent isolated-word recognition experiments under additive noise environments. The proposed method achieved outstanding improvement for speech recognition in environments with significant noise and was also effective in a range of feature representations.

• 대한전기학회논문지:시스템및제어부문D
• /
• 제51권6호
• /
• pp.267-271
• /
• 2002

The Hotelling transform is based on statistical properties of an image. The principal uses of this transform are in data compression. The basic concept of the Hotelling transform is that the choice of basis vectors pointing the direction of maximum variance of the data. This property can be used for rotation normalization. Many objects of interest in pattern recognition applications can be easily standardized by performing a rotation normalization that aligns the coordinate axes with the axes of maximum variance of the pixels in the object. However, this transform can not be used to rotation normalization of color images directly. In this paper, we propose a new method for rotation normalization of color images based on the Hotelling transform. The Hotelling transform is performed to calculate basis vectors of each channel. Then the summation of vectors of all channels are processed. Rotation normalization is performed using the result of summation of vectors. Experimental results showed the proposed method can be used for rotation normalization of color images effectively.

• Genomics & Informatics
• /
• 제5권3호
• /
• pp.118-123
• /
• 2007

The current existing literature offers little guidance on how to decide which method to use to analyze one-channel microarray measurements when dealing with large, grouped samples. Most previous methods have focused on two-channel data;therefore they can not be easily applied to one-channel microarray data. Thus, a more reliable method is required to determine an appropriate combination of individual basic processing steps for a given dataset in order to improve the validity of one-channel expression data analysis. We address key issues in evaluating the effectiveness of basic statistical processing steps of microarray data that can affect the final outcome of gene expression analysis without focusingon the intrinsic data underlying biological interpretation.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권1호
• /
• pp.168-182
• /
• 2020

Spoofing attacks, especially replay attacks, pose great security challenges to automatic speaker verification (ASV) systems. Current works on replay attacks detection primarily focused on either developing new features or improving classifier performance, ignoring the effects of feature variability, e.g., the channel variability. In this paper, we first establish a mathematical model for replay speech and introduce a method for eliminating the negative interference of the channel. Then a novel feature is proposed to detect the replay attacks. To further boost the detection performance, four post-processing methods using normalization techniques are investigated. We evaluate our proposed method on the ASVspoof 2017 dataset. The experimental results show that our approach outperforms the competing methods in terms of detection accuracy. More interestingly, we find that the proposed normalization strategy could also improve the performance of the existing algorithms.

• 방송공학회논문지
• /
• 제20권5호
• /
• pp.738-747
• /
• 2015

안개 영상에서는 색상정보와 테두리 정보가 줄어들기 때문에 사물의 식별이 어렵다. 안개 제거의 대표적인 알고리즘인 Dark Channel Prior(DCP)은 색 정보를 이용하여 안개의 전달량을 추정한 후 안개를 제거한다. 하지만 석양 또는 황사와 같이 안개에 영향을 미치는 요소가 영상에 포함되어있는 경우 안개 제거 후 특정 채널의 색상이 두드러지게 나타나는 문제점이 있다. 또한, RGB 채널이 모두 높은 값을 갖고 있는 사물이 포함된 영상의 경우 해당영역의 전달량이 오추정되는 문제점이 발생한다. 본 논문에서는 안개 영상의 백색 영역을 중심으로 개선된 색 정규화 방식을 적용한 후, 거리 정보를 바탕으로 오추정된 안개 영역을 보정하여 안개를 제거하는 방법을 제안한다. 제안하는 알고리즘을 통해 위와 같은 문제점을 보완하고 기존의 DCP 알고리즘보다 효과적으로 안개를 제거 할 수 있다.