• Proceedings of the Acoustical Society of Korea Conference
• /
• 1994.06a
• /
• pp.680-685
• /
• 1994

The acoustic phenomena in the actual sound systems involve a variety of compound problems. In this paper, the well-known Bayes' theorem is first employed and expanded into orthonormal and non-orthonomal series forms matched to the digital processing of lower and higher order statistical informations and the noisy observations. Proposed on-line algorithms of digital filter type are applied to the actual state estimation for a reverberation characteristics in a room under contamination of background noises.

• Journal of Korea Multimedia Society
• /
• v.16 no.8
• /
• pp.905-913
• /
• 2013

The Parallel Model Compensation Using Frequency-variant: FV-PMC for noise-robust speech recognition is a method to classify the noises, which are expected to be intermixed with input speech when recognized, into several groups of noises by setting average frequency variant as a threshold value; and to recognize the noises depending on the classified groups. This demonstrates the excellent performance considering noisy speech categorized as good using the standard threshold value. However, it also holds a problem to decrease the average speech recognition rate with regard to unclassified noisy speech, for it conducts the process of speech recognition, combined with noiseless model as in the existing PMC. To solve this problem, this paper suggests a enhanced method of recognition to prevent the unclassified through improving the extent of rating scales with use of maximum likelihood so that the noise groups, including input noisy speech, can be classified into more specific groups, which leads to improvement of the recognition rate. The findings from recognition experiments using Aurora 2.0 database showed the improved results compared with those from the method of the previous FV-PMC.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.41 no.3
• /
• pp.147-153
• /
• 2004

This paper proposes an efficient edge detection using Van der Waerden statistic in original and noisy images. An edge is where the intensity of an image moves from a low value to a high value or vice versa. We describe a nonparametric Wilcoxon test and a parametric T test based on statistical hypothesis testing for the detection of edges. We use the threshold determined by specifying significance level $\alpha$, while Bovik, Huang and Munson consider the range of possible values of test statistics for the threshold. From the experimental results of edge detection, the T and Wilcoxon method perform sensitively to the noisy image, while the proposed Waerden method is robust over both noisy and noise-free images under $\alpha$=0.0005. Comparison with our statistical test and Sobel, LoG, Canny operators shows that Waerden method perform more effectively in both noisy and noise-free images.

• Korean Journal of Psychosomatic Medicine
• /
• v.8 no.2
• /
• pp.181-190
• /
• 2000

This study describes the relationships between aircraft noise and stress and subjective noisy perception by comparing with two areas where the airport is adjacent or not. One area is Yipam-dong and Hak-dong expected to have louder aircraft noise because Kangnung Airport is near the area, and the other area is Songjung-dong as a control. For this study, we used telephone-survey which was targeted to two hundreds people of each area through the random sampling method. The design of the study is cross sectional study and the unit of analysis is a person. As the result of the multiple regression and logistic analysis, under the control of the other control variables, the both stress levels and subjective noisy perception of Yipam-dong and Hak-dong people significantly higher than that of Songjung-dong people and the people graduated middle school had significantly higher both stress levels and subjective noisy perception than the people graduated high school. In the future more studies have to be made that include concrete medical and psychiatric health problems.

• The Journal of the Acoustical Society of Korea
• /
• v.16 no.6
• /
• pp.94-99
• /
• 1997

This paper describes a technique of probabilistic spectral subtraction which uses the knowledge of both noise and speech so as to reduce automatic speech recognition errors in noisy environments. Spectral subtraction method estimates a noise prototype in non-speech intervals and the spectrum of clean speech is obtained from the spectrum of noisy speech by subtracting this noise prototype. Thus noise can not be suppressed effectively using a single noise prototype in case the characteristics of the noise prototype are different from those of the noise contained in input noisy speech. To modify such a drawback, multiple noise prototypes are used in probabilistic subtraction method. In this paper, the probabilistic characteristics of noise and the knowledge of speech which is embedded in hidden Markov models trained in clean environments are used to suppress noise. Futhermore, dynamic feature parameters are considered as well as static feature parameters for effective noise suppression. The proposed method reduced error rates in the recognition of 50 Korean words. The recognition rate was 86.25% with the probabilistic subtraction, 72.75% without any noise suppression method and 80.25% with spectral subtraction at SNR(Signal-to-Noise Ratio) 10 dB.

• The Journal of the Acoustical Society of Korea
• /
• v.16 no.6
• /
• pp.25-35
• /
• 1997

Most of speech analysis methods developed up to date are based on second order statistics, and one of the biggest drawback of these methods is that they show dramatical performance degradation in noisy environments. On the contrary, the methods using higher order statistics(HOS), which has the property of suppressing Gaussian noise, enable robust feature extraction in noisy environments. In this paper we propose a text-independent speaker identification system using higher order statistics and compare its performance with that using the conventional second-order-statistics-based method in both white and colored noise environments. The proposed speaker identification system is based on the vector quantization approach, and employs HOS-based voiced/unvoiced detector in order to extract feature parameters for voiced speech only, which has non-Gaussian distribution and is known to contain most of speaker-specific characteristics. Experimental results using 50 speaker's database show that higher-order-statistics-based method gives a better identificaiton performance than the conventional second-order-statistics-based method in noisy environments.

• 전자공학회논문지 IE
• /
• v.45 no.3
• /
• pp.13-19
• /
• 2008

In this paper, we propose a new IPC(Interlace to Progressive Conversion) method based on ELA(EDge Line based Average) interpolation using detecting the reliable edge direction. Existing ELA algorithms execute linear interpolation using edge direction without considering noises. In noisy images, these algorithms degrade quality because if interpolation based on the wrong edge direction. Out scheme is able to solve the problem of existing ELA algorithms in noisy images. First, filter a noisy pixel and estimate sizes of the noiseless orginal pixed and the noise, repectively. Then, considering the size of the noise, calculate weights of ELA and vertical interpolation. If noises exist after IPC, these could be eliminated by post filtering. The experimental results show that our proposed algorithm has about $1{\sim}2$ dB better performance than those of existing ELA algorithms.

• Journal of Korean Academy of Nursing
• /
• v.40 no.5
• /
• pp.736-746
• /
• 2010

Purpose: This study was done to identify the distribution and related factors for stage of change for wearing hearing protection devices (HPDs) by workers in environments with high noise. Predictors of Use of Hearing Protection Model and Trans-theoretical Model were tested. Methods: The participants were 755 workers from 20 noisy work places in Busan and Gyeongnam. Data were collected from January to April 2008 using self-administered questionnaires, and analyzed using multiple logistic regression. Results: There were significant differences in social mode (OR=1.35, 95% CI: 1.06-1.73) between precontemplation/contemplation and preparation stage, in males (OR=2.36, 95% CI: 1.24-4.51), workers with high school education or less (OR=1.39, 95% CI: 1.28-2.78), shift workers (OR=1.50, 95% CI: 1.02-2.21), workers who previously worked in noisy places (OR=1.39, 95% CI: 1.20-2.34), and workers who had previous hearing examinations (OR=1.89, 95% CI: 1.25-2.85), in the social model (OR=1.59, 95% CI: 1.42-1.78), and self-efficacy (OR=1.05, 95% CI: 1.02-1.08) between workers in preparation and action stages, in length of time working in noisy work places (OR=2.26, 95% CI: 1.17-4.39), social model (OR=1.66, 95% CI: 1.33-2.08), and perceived benefit (OR=0.95, 95% CI: 0.93-0.97) between action and maintenance stage. Conclusion: Social model was a common factor showing differences between two adjacent stages for wearing HPDs. The results provide data for developing programs to encourage workers to wear HPDs and application of these programs in work settings.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.17 no.7
• /
• pp.1571-1576
• /
• 2013

This paper proposes an improvement algorithm of signal-to-noise ratios (SNRs) for speech signals under noisy environments. The proposed algorithm first estimates the SNRs in a low SNR, mid SNR and high SNR areas, in order to improve the SNRs in the speech signal from background noise, such as white noise and car noise. Thereafter, this algorithm subtracts the noise signal from the noisy speech signal at each bands using a spectrum sharpening method. In the experiment, good signal-to-noise ratios (SNR) are obtained for white noise and car noise compared with a conventional spectral subtraction method. From the experiment results, the maximal improvement in the output SNR results was approximately 4.2 dB and 3.7 dB better for white noise and car noise compared with the results of the spectral subtraction method, in the background noisy environment, respectively.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.41 no.3
• /
• pp.67-77
• /
• 2004

The recent research has been focused on fusion of audio and visual features for reliable speech recognition in noisy environments. In this paper, we propose a neural network based model of robust speech recognition by integrating audio, visual, and contextual information. Bimodal Neural Network(BMNN) is a multi-layer perception of 4 layers, each of which performs a certain level of abstraction of input features. In BMNN the third layer combines audio md visual features of speech to compensate loss of audio information caused by noise. In order to improve the accuracy of speech recognition in noisy environments, we also propose a post-processing based on contextual information which are sequential patterns of words spoken by a user. Our experimental results show that our model outperforms any single mode models. Particularly, when we use the contextual information, we can obtain over 90% recognition accuracy even in noisy environments, which is a significant improvement compared with the state of art in speech recognition. Our research demonstrates that diverse sources of information need to be integrated to improve the accuracy of speech recognition particularly in noisy environments.