• MALSORI
• /
• no.52
• /
• pp.145-160
• /
• 2004

In this paper, a new noise compensation method based on the eigenvoice framework in feature space is proposed to reduce the mismatch between training and testing environments. The difference between clean and noisy environments is represented by the linear combination of K eigenvectors that represent the variation among environments. In the proposed method, the performance improvement of speech recognition systems is largely affected by how to construct the noisy models and the bias vector set. In this paper, two methods, the one based on MAP adaptation method and the other using stereo DB, are proposed to construct the noisy models. In experiments using Aurora 2 DB, we obtained 44.86% relative improvement with eigen-environment method in comparison with baseline system. Especially, in clean condition training mode, our proposed method yielded 66.74% relative improvement, which is better performance than several methods previously proposed in Aurora project.

• The Journal of the Acoustical Society of Korea
• /
• v.31 no.2
• /
• pp.87-92
• /
• 2012

In this paper, a robust feature compensation method to deal with the environmental mismatch is proposed. The proposed method applies energy based weights according to the degree of speech presence to the Mean subtraction, Variance normalization, and ARMA filtering (MVA) processing. The weights are further smoothed by the moving average and maximum filters. The proposed feature compensation algorithm is evaluated on AURORA 2 task and distant talking experiment using the robot platform, and we obtain error rate reduction of 14.4 % and 44.9 % by using the proposed algorithm comparing with MVA processing on AURORA 2 task and distant talking experiment, respectively.

• International Journal of Advanced Culture Technology
• /
• v.7 no.4
• /
• pp.194-207
• /
• 2019

This paper is the author's personal experience and interpretation as a student whilst participating in Professor Kyung-Sook Shin's English Literature graduate course, "Literature and Technology II: Feminisms and Digital Humanities," during the 2019 spring semester at Yonsei University, South Korea. Exploring the intersections of literary feminist theory and digital humanities, this paper examines not only the content, but also the methodology and political effects of collaboratively digitally annotating Elizabeth Barrett Browning's epic novel/poem, Aurora Leigh (1856) through the medium, Google Docs. In particular, this paper observes the students' interaction with the digital tools and literature-related pedagogy in two main parts. First, the democratic political nature of classroom culture when creating a new language/code during annotation. Second, the coexistence of cyberspace and the physical classroom space and its effect on time, specifically in the archival of the past, influencing of the future, and the splitting into the present multiverse. From a student's perspective in digital literary annotation, this paper shows that technology could become a way to decolonize and reprogram education to be more inclusive and collaborative.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.8 no.4
• /
• pp.555-560
• /
• 2013

This paper first proposes a classification algorithm which detects a voiced, unvoiced, and silence signal using distance measure, logarithm power and root mean square methods at each frame, then a spectrum subtraction algorithm based on a subband filter. The proposed algorithm subtracts spectrums of white noise and street noise from noisy signal based on the subband filter at each frame. In this experiment, experimental results of the proposed spectrum subtraction algorithm demonstrate using the speech and noise data of Aurora-2 database. Based on measuring the speech-to-noise ratio (SNR), experiments confirm that the proposed algorithm is effective for the speech by contaminated the noise. From the experiments, the improvement in the output SNR values was approximately 2.1 dB and 1.91 dB better for white noise and street noise, respectively.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.46 no.2
• /
• pp.120-126
• /
• 2009

In order to reduce the noise quantity in a conversation under the noisy environment it is necessary for the signal processing system to process adaptively according to the noise quantity in order to enhance the performance. Therefore this paper presents a recognition method for noise quantity by linear predictive coefficient using a three layered neural network, which is trained using three kinds of speech that is degraded by various background noises. The performance of the proposed method for the noise quantity was evaluated based on the recognition rates for various noises. In the experiment, the average values of the recognition results were 98.4% or more for such noise using Aurora2 database.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.17 no.8
• /
• pp.1769-1776
• /
• 2013

In this paper, we proposed the modified YL(Yi and Loizou) algorithm, using a VAD(voice activity detector) for enhancing speech corrupted by colored noise. The performance of the proposed algorithm has been compared to the YL algorithm and LS(Lee and Son, etc.) algorithm by computer simulation. The colored noises used in the experiment were a car noise and multi-talker babble from the AURORA data base and the used voices from the TIMIT data base. It is confirmed that the proposed algorithm shows better performance from SNR(signal to noise ratio) and SSD(speech spectral distortion) viewpoint over the previous two approach.

• The Journal of the Acoustical Society of Korea
• /
• v.39 no.1
• /
• pp.64-68
• /
• 2020

In a conventional speech recognition system using Gaussian Mixture Model-Hidden Markov Model (GMM-HMM), the cepstral feature normalization method based on pole filtering was effective in improving the performance of recognition of short utterances in noisy environments. In this paper, the usefulness of this method for the state-of-the-art speech recognition system using Deep Neural Network (DNN) is examined. Experimental results on AURORA 2 DB show that the cepstral mean and variance normalization based on pole filtering improves the recognition performance of very short utterances compared to that without pole filtering, especially when there is a large mismatch between the training and test conditions.

• Bulletin of the Korean Space Science Society
• /
• 2008.10a
• /
• pp.31.1-31.1
• /
• 2008

While some observations in the geomagnetic tail region supported electrons could be accelerated by reconnection processes, we still need more observation data to confirm electron acceleration in this region. Because most acceleration processes accompany strong pitch angle diffusion, if the electrons were accelerated in this region, strong energetic electron precipitation should be observed near earth on aurora oval. Even though there are several low altitude satellites observing electron precipitation, intense and small scale precipitation events have not been identified successfully. In this presentation, we will show an observation of strong energetic electron precipitation that might be analyzed by relativistic electron acceleration in the confined region. This event was observed by low altitude Korean STSAT-1, where intense several hundred keV electron precipitation was seen simultaneously with 10 keV electrons during storm time. In addition, we observed large magnetic field fluctuations and an ionospheric plasma depletion with FUV aurora emissions. Our observation implies relativistic electrons can be generated in the small area where Fermi acceleration might work.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.37 no.6C
• /
• pp.492-496
• /
• 2012

This paper presents a decoding technique for speech recognition using uncertainty information from feature compensation method to improve the speech recognition performance in the low SNR condition. Traditional feature compensation algorithms have difficulty in estimating clean feature parameters in adverse environment. Those algorithms focus on the point estimation of desired features. The point estimation of feature compensation method degrades speech recognition performance when incorrectly estimated features enter into the decoder of speech recognition. In this paper, we apply the uncertainty information from well-known feature compensation method, such as IMM, to the recognition engine. Applied technique shows better performance in the Aurora-2 DB.

• ETRI Journal
• /
• v.30 no.5
• /
• pp.753-755
• /
• 2008

In this letter, we propose a new histogram equalization technique for feature compensation in speech recognition under noisy environments. The proposed approach combines a signal-to-noise-ratio-dependent feature reconstruction method and the class histogram equalization technique to effectively reduce the acoustic mismatch present in noisy speech features. Experimental results from the Aurora 2 task confirm the superiority of the proposed approach for acoustic feature compensation.