• The Journal of the Acoustical Society of Korea
• /
• v.22 no.4
• /
• pp.299-307
• /
• 2003

There have been many studies on speaker normalization which aims to minimize the effects of speaker's vocal tract length on the recognition performance of the speaker independent speech recognition system. In this paper, we propose a simple vector quantizer based linear warping speaker normalization method based on the observation that the vector quantizer can be successfully used for speaker verification. For this purpose, we firstly generate an optimal codebook which will be used as the basis of the speaker normalization, and then the warping factor of the unknown speaker will be extracted by comparing the feature vectors and the codebook. Finally, the extracted warping factor is used to linearly warp the Mel scale filter bank adopted in the course of MFCC calculation. To test the performance of the proposed method, a series of recognition experiments are conducted on discrete HMM with thirteen mono-syllabic Korean number utterances. The results showed that about 29% of word error rate can be reduced, and that the proposed warping factor extraction method is useful due to its simplicity compared to other line search warping methods.

• Journal of the Society of Naval Architects of Korea
• /
• v.47 no.6
• /
• pp.757-762
• /
• 2010

Korean fishing boats have had appropriate hull forms for the safety, stability and convenience of fishing ability. However, Korean fishermen are recently concerned about the resistance performance and speed of Korean fishing boats, because the prices of fuel oil are gradually risen, also the exhausting of fish resources and the demand of high speed fishing boats are increased. Therefore, the necessity of the study on the improvement of resistance performance for Korean small coastal fishing boats is gradually increased. This study compares the hull form characteristics of Korean fishing boats with those of Japanese fishing boats, and the hull form of a representative Korean fishing boat is modified. From the modification of the hull form parameters for the Korean fishing boat, the improvement of resistance performances is evaluated. Moreover, the increase of resistance performances is also achieved from the modification of local characteristics for the hull form of the Korean fishing boat. A computational method and ship model tests in towing tank are used for the conformations of the improvement of resistance performance.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.16 no.3
• /
• pp.401-406
• /
• 2021

Recently, mean-teacher models based on convolutional recurrent neural networks are popularly used in audio event detection. The mean-teacher model is an architecture that consists of two parallel CRNNs and it is possible to train them effectively on the weakly-labelled and unlabeled audio data by using the consistency learning metric at the output of the two neural networks. In this study, we tried to improve the performance of the mean-teacher model by using additional derivative features of the log-mel spectrum. In the audio event detection experiments using the training and test data from the Task 4 of the DCASE 2018/2019 Challenges, we could obtain maximally a 8.1% relative decrease in the ER(Error Rate) in the mean-teacher model using proposed derivative features.

• The Journal of the Acoustical Society of Korea
• /
• v.38 no.4
• /
• pp.433-441
• /
• 2019

In this paper, a method of estimating speech parameters for ultrasonic Doppler signals reflected from the articulatory muscles using LSTM (Long Short Term Memory) RNN (Recurrent Neural Networks) was introduced and compared with the method using MLP (Multi-Layer Perceptrons). LSTM RNN were used to estimate the Fourier transform coefficients of speech signals from the ultrasonic Doppler signals. The log energy value of the Mel frequency band and the Fourier transform coefficients, which were extracted respectively from the ultrasonic Doppler signal and the speech signal, were used as the input and reference for training LSTM RNN. The performance of LSTM RNN and MLP was evaluated and compared by experiments using test data, and the RMSE (Root Mean Squared Error) was used as a measure. The RMSE of each experiment was 0.5810 and 0.7380, respectively. The difference was about 0.1570, so that it confirmed that the performance of the method using the LSTM RNN was better.

• Dementia and Neurocognitive Disorders
• /
• v.23 no.1
• /
• pp.1-10
• /
• 2024

Background and Purpose: Voice, reflecting cerebral functions, holds potential for analyzing and understanding brain function, especially in the context of cognitive impairment (CI) and Alzheimer's disease (AD). This study used voice data to distinguish between normal cognition and CI or Alzheimer's disease dementia (ADD). Methods: This study enrolled 3 groups of subjects: 1) 52 subjects with subjective cognitive decline; 2) 110 subjects with mild CI; and 3) 59 subjects with ADD. Voice features were extracted using Mel-frequency cepstral coefficients and Chroma. Results: A deep neural network (DNN) model showed promising performance, with an accuracy of roughly 81% in 10 trials in predicting ADD, which increased to an average value of about 82.0%±1.6% when evaluated against unseen test dataset. Conclusions: Although results did not demonstrate the level of accuracy necessary for a definitive clinical tool, they provided a compelling proof-of-concept for the potential use of voice data in cognitive status assessment. DNN algorithms using voice offer a promising approach to early detection of AD. They could improve the accuracy and accessibility of diagnosis, ultimately leading to better outcomes for patients.

• Korean Journal of Breeding Science
• /
• v.42 no.4
• /
• pp.397-406
• /
• 2010

Field screening method has been commonly used for purity test of $F_1$ hybrid seeds in melon and oriental melon. However, as this method takes a lot of time and cost, molecular marker-based purity test is necessary. To develop molecular markers for purity test, thirty pairs of SNP (single nucleotide polymorphism) primers were obtained from melon EST sequences, and 10 polymorphic markers showing HRM (high resolution melting) polymorphisms between parents of two melon cultivars and one oriental melon cultivar were selected. Blind tests were performed to validate usefulness of the selected markers for purity test. Blind test results showed that HRM genotypes were matched with the expected identity of individual sample, $F_1$ hybrid, male or female parents. Three HRM-based SNP markers were converted to CAPS markers for general use which is favor to breeders. We expect that SNP markers developed in this study will be useful for purity test of $F_1$ hybrid seeds in melon and oriental melon.

• Journal of Life Science
• /
• v.26 no.6
• /
• pp.633-639
• /
• 2016

Biomaterials including polymer, metal, ceramic, and composite have been widely applied for medical uses as medical fibers, artificial blood vessels, artificial joints, implants, soft tissue, and plastic surgery materials owing to their physicochemical properties. However, the biocompatibility and toxicity for film- and plate-form biomaterials is difficult to measure in mammalian cells because there is no appropriate incubation system. To solve these problems, we developed a novel mammalian cell culture system consisting of a silicone ring, top panel, and bottom panel and we applied two polymer films (PF) and one metal plate (MP). This system was based on the principal of sandwiching a test sample between the top panel and the bottom panel. Following the assembly of the culture system, SK-MEL-2 cells were seeded onto Styela Clava Tunic (SCT)-PF, NaHCO₃-added SCT (SCTN)-PF, and magnesium MP (MMP) and incubated at 37℃ for 24 hr and 48 hr. An MTT assay revealed that cell viability was maintained at a normal level in the SCT-PF culture group at 24 or 48 hr, although it rapidly decreased in the SCTN-PF culture group at 48 hr. Furthermore, the cell viability in the MMP culture group was very similar to that of the control group after incubation for 24 hr and 48 hr. Together, these results suggest the sandwich-type mammalian culture system developed here has the potential for the evaluation of the biocompatibility and toxicity of cells against PF- and MP-form biomaterials.

• The Journal of the Acoustical Society of Korea
• /
• v.23 no.2
• /
• pp.180-186
• /
• 2004

The presence of noise in speech signals degrades the performance of recognition systems in which there are mismatches between the training and test environments. To make a speech recognizer robust, it is necessary to compensate these mismatches. In this paper, we studied about an improvement of stochastic feature extraction based on band-SNR for robust speech recognition. At first, we proposed a modified version of the multi-band spectral subtraction (MSS) method which adjusts the subtraction level of noise spectrum according to band-SNR. In the proposed method referred as M-MSS, a noise normalization factor was newly introduced to finely control the over-estimation factor depending on the band-SNR. Also, we modified the architecture of the stochastic feature extraction (SFE) method. We could get a better performance when the spectral subtraction was applied in the power spectrum domain than in the mel-scale domain. This method is denoted as M-SFE. Last, we applied the M-MSS method to the modified stochastic feature extraction structure, which is denoted as the MMSS-MSFE method. The proposed methods were evaluated on isolated word recognition under various noise environments. The average error rates of the M-MSS, M-SFE, and MMSS-MSFE methods over the ordinary spectral subtraction (SS) method were reduced by 18.6%, 15.1%, and 33.9%, respectively. From these results, we can conclude that the proposed methods provide good candidates for robust feature extraction in the noisy speech recognition.