• 대한전자공학회논문지SP
• /
• 제40권5호
• /
• pp.371-377
• /
• 2003

본 논문에서 음성신호를 사용하여 인간의 감정를 인식하기 위한 특징 파라메터 비교에 관하여 연구하였다. 이를 위하여 여러 가지 감정 상태에 따라 분류된 한국어 음성 데이터 베이스를 이용하여 얻어진 음성 신호의 피치와 에너지의 평균, 표준편차와 최대 값 등 통계적인 정보 나타내는 파라메터와 음소의 특성을 나타내는 MFCC 파라메터가 사용되었다. 파라메터들의 성능을 평가하기 위하여 문장 및 화자 독립 감정 인식 시스템을 구현하여 인식 실험을 수행하였다. 성능 평가를 위한 실험에서는 운율적 특징으로 피치와 에너지와 각각의 미분 값을 사용하였고, 음소의 특성을 나타내는 특징으로 MFCC와 그 미분 값을 사용하였다. 벡터 양자화 방법을 사용한 화자 및 문장 독립 인식 시스템을 사용한 실험 결과에서 MFCC와 델타 MFCC를 사용한 경우가 피치와 에너지를 사용한 방법보다 우수한 성능을 나타내었다.

• 말소리와 음성과학
• /
• 제1권2호
• /
• pp.9-14
• /
• 2009

Speech enhancement methods can be generally classified into a few categories and they have been usually compared with each other in terms of speech quality. For the successful use of speech enhancement methods in speech recognition systems, performance comparisons in terms of speech recognition accuracy are necessary. In this paper, we compared the speech recognition performance of some of the representative speech enhancement algorithms which are popularly cited in the literature and used widely. We also compared the performance of speech enhancement methods with other noise robust speech recognition methods like PMC to verify the usefulness of speech enhancement approaches in noise robust speech recognition systems.

• Journal of Information Processing Systems
• /
• 제8권3호
• /
• pp.483-494
• /
• 2012

Many recent events, such as terrorist attacks, exposed defects in most sophisticated security systems. Therefore, it is necessary to improve security data systems based on the body or behavioral characteristics, often called biometrics. Together with the growing interest in the development of human and computer interface and biometric identification, human face recognition has become an active research area. Face recognition appears to offer several advantages over other biometric methods. Nowadays, Principal Component Analysis (PCA) has been widely adopted for the face recognition algorithm. Yet still, PCA has limitations such as poor discriminatory power and large computational load. This paper proposes a novel algorithm for face recognition using a mid band frequency component of partial information which is used for PCA representation. Because the human face has even symmetry, half of a face is sufficient for face recognition. This partial information saves storage and computation time. In comparison with the traditional use of PCA, the proposed method gives better recognition accuracy and discriminatory power. Furthermore, the proposed method reduces the computational load and storage significantly.

• International Journal of Advanced Culture Technology
• /
• 제6권3호
• /
• pp.173-177
• /
• 2018

Masks are generated by adding different fonts of learning data characters in pixel unit, and pixel values belonging to each of the masks are divided into 3 groups. Using the directional expansion operators, we expand the text area of the test data character into 4 diagonal directions in order to create the boundary areas to distinguish it from the background area. A mask with a minimum average discordance is selected as the final recognition result by calculating the degree of discordance between the expanded test data and the masks. Image comparison using directional expansion operations more accurately recognizes test data through 4 subdivided recognition processes. It is also possible to expand the ranges of 3 groups of pixel values of masks more evenly such that new fonts can easily be added to the given learning data.

• 음성과학
• /
• 제10권1호
• /
• pp.27-39
• /
• 2003

This study attempts to verify the reliability of automatically generated segment labels as compared to those obtained by conventional labelling by hand. First of all, an autolabeller is constructed using the standard HMM speech recognition technique. For evaluation, we compare the automatically generated labels with manually annotated labels for the same speech data. The comparison is performed by calculating the temporal difference between an autolabel boundary and its corresponding hand label boundary. When the mismatched duration between two labels falls within 10 msec, we consider the autolabel as correct. The results suggest that overall 78% of autolabels are correctly obtained. It is found that the boundary of obstruents is better aligned than that of sonorants and vowels. In case of stop sound classes, strong stops in manner-of-articulation wise and velar stops in place-of-articulation wise show better performance in boundary alignment. The result suggests that more phone-specific consideration is necessary to improve autosegmentation performance.

• 대한지역사회영양학회지
• /
• 제8권1호
• /
• pp.71-82
• /
• 2003

In this study, hygiene knowledge and recognition on job performance levels of foodservice employees at different degrees of HACCP application had been compared. As for the comparison of hygiene knowledge for HACCP application, foodservice employees at "Appointed" showed significantly higher HACCP knowledge (p < 0.01) . As for recognition on job performance levels, this status also assured the significant difference between the comparison groups (p < 0.01) : "Appointed" showed the highest point of 4.24 while "Voluntary applying" and "Non-applying" showed 3.39 and 2.53 respectively. The "Appointed" group showed the well performance of most of the surveyed items. "Voluntary applying" group showed unsatisfactory performance in various log recordings and some part of surveyed items. As for "Non-applying", many items were performed under the average score as they do not apply HACCP.ems were performed under the average score as they do not apply HACCP.

• 전자공학회논문지B
• /
• 제28B권12호
• /
• pp.68-74
• /
• 1991

This paper proposes a HMM model based on fuzzy, as a method on the speech recognition of speaker-independent. In this recognition method, multi-observation sequences which give proper probabilities by fuzzy rule according to order of short distance from VQ codebook are obtained. Thereafter, the HMM model using this multi-observation sequences is generated, and in case of recognition, a word that has the most highest probability is selected as a recognized word. The vocabularies for recognition experiment are 146 DDD are names, and the feature parameter is 10S0thT LPC cepstrum coefficients. Besides the speech recognition experiments of proposed model, for comparison with it, we perform the experiments by DP, MSVQ and general HMM under same condition and data. Through the experiment results, it is proved that HMM model using fuzzy proposed in this paper is superior to DP method, MSVQ and general HMM model in recognition rate and computational time.

• 음성과학
• /
• 제1권
• /
• pp.277-283
• /
• 1997

In this paper, we present a robust recognition algorithm based on the Wiener filtering method as a research tool to develop the Korean Speech recognition system. We especially used Wiener filtering method in cepstrum-domain, because the method in frequency-domain is computationally expensive and complex. Evaluation of the effectiveness of this method has been conducted in speaker-independent isolated Korean digit recognition tasks using discrete HMM speech recognition systems. In these tasks, we used 12th order weighted cepstral as a feature vector and added computer simulated white gaussian noise of different levels to clean speech signals for recognition experiments under noisy conditions. Experimental results show that the presented algorithm can provide an improvement in recognition of as much as from $5\%\;to\;\20\%$ in comparison to spectral subtraction method.

• 말소리와 음성과학
• /
• 제2권4호
• /
• pp.153-161
• /
• 2010

Many isolated word recognition systems may generate non-similar words for recognition candidates because they use only acoustic information. The previous study [1,2] investigated several techniques which can exclude non-similar words from N-best candidate words by applying Levenstein distance measure. This paper discusses the various improving techniques of removing non-similar recognition results. The mentioned methods include comparison penalties or weights, phone accuracy based on confusion information, weights candidates by ranking order and partial comparisons. Through experimental results, it is found that some proposed method keeps more accurate recognition results than the previous method's results.

• International Journal of Advanced Culture Technology
• /
• 제7권3호
• /
• pp.134-142
• /
• 2019

As the 4th industry has been widely distributed, there is a need for a process of real-time image recognition in various fields such as identification of company employees, security maintenance, and development of military weapons. Therefore, in this paper, we will propose an algorithm that effectively recognizes a test object by comparing it with the DB model. The proposed object recognition system first expresses the outline of the test object as a set of vertices with the distances of predefined length or more. Then, the degree of matching of the structures of the two objects is calculated by examining the distances to the outline of the DB model from the vertices constituting the test object. Because the proposed recognition algorithm uses the outline of the object, the recognition process is easy to understand, simple to implement, and a satisfactory recognition result is obtained.