• Speech Sciences
• /
• v.14 no.4
• /
• pp.145-156
• /
• 2007

The performance of speaker identification is severely degraded in noisy environments. Recently probability weighting method based on observation membership was proposed for overcoming the noise problem[1]. In the paper[1] the observation confidence was calculated from SNR with sigmoid function. However, estimating SNR needs additive calculation amount and estimated SNR is corrupted in dynamic noisy environments. In this paper we propose estimation methods of the observation confidence based on score-based reliabilities (SBR) of entropy and dispersion measures. Generally SBRs are obtained from speaker models' probabilities. The proposed methods are evaluated with ETRI speaker recognition DB. We compared the performances of the proposed methods with those in [1][8]. The experimental results show that the proposed methods can be successfully applied for the case where SNR is not available.

• Speech Sciences
• /
• v.15 no.2
• /
• pp.79-88
• /
• 2008

Recently demands for the speech technology-based products targeted for the mobile terminals such as cellular phones and PDA are rapidly increasing. And voting-based speaker identification algorithm is known to have a good performance in the mobile environment, since it works well with small amount of speaker training data. In this paper, we proposed a method to improve the performance of this voting based speaker identification system by using the observation confidence value which is derived from the function of SNR each frame. The proposed method is evaluated with ETRI cellular phone DB which is made for the speaker recognition task. The experimental results show that the proposed method has better performance of 2-3% identification rate than the conventional GMM method.

• MALSORI
• /
• no.64
• /
• pp.121-135
• /
• 2007

Speech signal is distorted by channel characteristics or additive noise and then the performances of speaker or speech recognition are severely degraded. To cope with the noise problem, we propose a modified HMM decoder algorithm using SNR-based observation confidence, which was successfully applied for GMM in speaker identification task. The modification is done by weighting observation probabilities with reliability values obtained from SNR. Also, we apply PSO (particle swarm optimization) method to the confidence function for maximizing the speaker identification performance. To evaluate our proposed method, we used the ETRI database for speaker recognition. The experimental results showed that the performance was definitely enhanced with the modified HMM decoder algorithm.

• The Korean Journal of Emergency Medical Services
• /
• v.22 no.2
• /
• pp.67-77
• /
• 2018

Purpose: This study was conducted to identify the factors influencing performance confidence by knowledge and experience of respiratory-assistant therapy in paramedic students. Methods: A descriptive survey study used a convenience sample comprising 387 students from September 15 to October 31, 2016. Data were analyzed using descriptive statistics, t-test, ANOVA, Pearson's correlation coefficient, and regression analysis. Results: Knowledge was statistically significant by college(3-year), acquired certification, and observation experience in respiratory-assistant therapy. Performance confidence was statistically significant by sex, academic grade($3.0{\leq}$), acquired certification, and observation and performance experience in respiratory-assistant therapy. There was a significant correlation between knowledge and acquired certification, clinical practice, and major satisfaction. There was a significant correlation between performance confidence and acquired certification, clinical practice, major satisfaction, and knowledge. Regression analysis revealed that performance confidence accounted for 22.9% of the variance by academic grade(3.0~3.99), acquired certification, and major satisfaction. Conclusion: For paramedic students in clinical practice and in-school classes, it is important to improve performance confidence by combining training instruction with experience and education of respiratory-assistant therapy.

• Communications for Statistical Applications and Methods
• /
• v.16 no.4
• /
• pp.675-686
• /
• 2009

In this paper, it will be assumed that there are two distinct populations which are multivariate normal with equal covariance matrix. We also assume that the two populations are equally likely and the costs of misclassification are equal. The classification rule depends on the situation when the training samples include missing values or not. We consider the bootstrap confidence intervals for classification error rate when a block of observation is missing.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.4
• /
• pp.1824-1845
• /
• 2016

The GMM is a conventional approach which has been recently applied in many face recognition studies. However, the question about how to deal with illumination changes while ensuring high performance is still a challenge, especially with real-world databases. In this paper, we propose a Visual Observation Confidence (VOC) measure for robust face recognition for illumination changes. Our VOC value is a combined confidence value of three measurements: Flatness Measure (FM), Centrality Measure (CM), and Illumination Normality Measure (IM). While FM measures the discrimination ability of one face, IM represents the degree of illumination impact on that face. In addition, we introduce CM as a centrality measure to help FM to reduce some of the errors from unnecessary areas such as the hair, neck or background. The VOC then accompanies the feature vectors in the EM process to estimate the optimal models by modified-GMM training. In the experiments, we introduce a real-world database, called KoFace, besides applying some public databases such as the Yale and the ORL database. The KoFace database is composed of 106 face subjects under diverse illumination effects including shadows and highlights. The results show that our proposed approach gives a higher Face Recognition Rate (FRR) than the GMM baseline for indoor and outdoor datasets in the real-world KoFace database (94% and 85%, respectively) and in ORL, Yale databases (97% and 100% respectively).

• Speech Sciences
• /
• v.14 no.1
• /
• pp.163-174
• /
• 2007

All observation has uncertainty due to noise or channel characteristics. This uncertainty should be counted in the modeling of observation. In this paper we propose a modified optimization object function of a GMM training considering inexact observation. The object function is modified by introducing the concept of observation confidence as a weighting factor of probabilities. The optimization of the proposed criterion is solved using a common EM algorithm. To verify the proposed method we apply it to the speaker recognition domain. The experimental results of text-independent speaker identification with VidTimit DB show that the error rate is reduced from 14.8% to 11.7% by the modified GMM training.

• Journal of Korean Elementary Science Education
• /
• v.34 no.4
• /
• pp.458-475
• /
• 2015

The purpose of this study was to analyze the responses of pre-service elementary teachers to the plant learning using plant-observation activities and observation journal writings. Thirty pre-service teachers participated in this study. Before and after practicing plant observation and observation journal writing for 11 weeks, pre-service teachers' interest to plants, attitude to plant-observation teaching, and plant-observation teaching efficacy were examined using a questionnaire. Also, the observation journals, reflective journal writings, and interviews were analyzed to investigate their qualitative change. In addition, the pre-service teachers were classified by multi-dimensional scaling and cluster analysis with respect to attitude to plant-observation teaching and observation journal writing, and teaching efficacy. While the pre-service teachers showed significant increase of the interest to plants and plant-observation teaching efficacy, and positive change of the attitude to plant-observation teaching, there was a little difference between gender with more positive effects on females. The positive attitude changes of pre-service teachers included positive recognition about plants and the diversity, increased interest and curiosity about plants, careful observation about nearby plants, and independent learning and positive inquiry with plants. In terms of teaching efficacy, they showed positive change such as enjoyment and satisfaction with plant-observation, confidence as elementary teachers in the future, and recognition about the importance of elementary teachers for plant education. However, the changes were different among the groups classified by attitude to plant-observation teaching and observation journal writings, and plant-observation teaching efficacy.

• The Journal of Korean Physical Therapy
• /
• v.32 no.6
• /
• pp.365-371
• /
• 2020

Objective: To investigate the action observation effects of functional electrical stimulation (FES) on the communication between motor cortex and muscle through corticomuscular coherence (CMC) analysis. Methods: Electroencephalogram (EEG) and electromyogram (EMG) of 27 healthy, nonathlete subjects were measured during action observation, FES, and action observation with FES, which lasted for 7sper session for 10 times. All trials were repeated for 30 times. Simultaneously measured EEG raw data and rectified EMG signals were used to calculate CMC. Only confidence limit values above 0.0306 were used for analysis. CMC was divided into three frequency domains, andthe grand average coherence and peak coherence were computed. Repeated ANOVA was performed to analyze the coherence value difference for each condition's frequency band. Results: CMC showed significant differences in peak coherence and average coherence between the conditions (p<0.05). Action observation application with FES in all frequency band showed the highest peak and average coherence value. Conclusions: The results of this study are assumed to be the combination of increased eccentric information transfer from the sensorymotor cortex by action observation and an increased in concentric sensory input from the peripheral by the FES, suggesting that these are reflecting the sensorimotor integration process.

• Journal of Korean Academy of Nursing
• /
• v.37 no.1
• /
• pp.125-134
• /
• 2007

Purpose: The study aimed to identify the effects of a newborn care education program on newborn care confidence and behavioral accuracy of primiparas in a postpartum care center. Method: This was a quasi-experimental study with a non-equivalent control group non-synchronized design. Subjects were 29 primiparas selected from a postpartum care center in D and P hospitals in Ulsan. The intervention was a newborn care education program which was composed of education with a booklet, demonstration, watching a video, and verbal encouragement. The outcome variables were newborn care confidence and behavioral accuracy. Data was collected from Feb. 1 to Mar 22 in 2005 with self-administered questionnaires and observation by researchers. Data was analyzed using the $X^2-test$, Fisher's exact test, t-test and ANCOVA. Result: The experimental group showed significantly higher score changes between the pre-test and post-test in confidence and behavioral accuracy than the control group. Conclusion: The newborn care education program showed positive effects on newborn care confidence and behavioral accuracy of primiparas in a postpartum care center. Therefore, we recommend that this program should be applied in postpartum care centers.