• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2008년도 심포지엄 논문집 정보 및 제어부문
• /
• pp.99-100
• /
• 2008

This paper proposes a Gaussian Mixture Model (GMM) - based speech emotion recognition methods using four feature parameters; 1) Fast Fourier Transform(FFT) spectral entropy, 2) delta FFT spectral entropy, 3) Mel-frequency Filter Bank (MFB) spectral entropy, and 4) delta MFB spectral entropy. In addition, we use four emotions in a speech database including anger, sadness, happiness, and neutrality. We perform speech emotion recognition experiments using each pre-defined emotion and gender. The experimental results show that the proposed emotion recognition using FFT spectral-based entropy and MFB spectral-based entropy performs better than existing emotion recognition based on GMM using energy, Zero Crossing Rate (ZCR), Linear Prediction Coefficient (LPC), and pitch parameters. In experimental Results, we attained a maximum recognition rate of 75.1% when we used MFB spectral entropy and delta MFB spectral entropy.

• 융합신호처리학회논문지
• /
• 제12권4호
• /
• pp.228-236
• /
• 2011

Emotion recognition plays an important role in the research area of human-computer interaction, and it allows a more natural and more human-like communication between humans and computer. Most of previous work on emotion recognition focused on extracting emotions from face, speech or EEG information separately. Therefore, a novel approach is presented in this paper, including face, speech and EEG, to recognize the human emotion. The individual matching scores obtained from face, speech, and EEG are combined using a weighted-summation operation, and the fused-score is utilized to classify the human emotion. In the experiment results, the proposed approach gives an improvement of more than 18.64% when compared to the most successful unimodal approach, and also provides better performance compared to approaches integrating two modalities each other. From these results, we confirmed that the proposed approach achieved a significant performance improvement and the proposed method was very effective.

• ETRI Journal
• /
• 제44권3호
• /
• pp.462-475
• /
• 2022

Although researchers have proposed numerous techniques for speech emotion recognition, its performance remains unsatisfactory in many application scenarios. In this study, we propose a speech emotion recognition model based on a genetic algorithm (GA)-decision tree (DT) fusion of deep and acoustic features. To more comprehensively express speech emotional information, first, frame-level deep and acoustic features are extracted from a speech signal. Next, five kinds of statistic variables of these features are calculated to obtain utterance-level features. The Fisher feature selection criterion is employed to select high-performance features, removing redundant information. In the feature fusion stage, the GA is is used to adaptively search for the best feature fusion weight. Finally, using the fused feature, the proposed speech emotion recognition model based on a DT support vector machine model is realized. Experimental results on the Berlin speech emotion database and the Chinese emotion speech database indicate that the proposed model outperforms an average weight fusion method.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제4권1호
• /
• pp.18-22
• /
• 2004

It is difficult to understand a person's emotion, since it is subjective and vague. Therefore, we are proposing a method by which will effectively classify human emotions into two types (that is, single emotion and composition emotion). To verify validity of te proposed method, we conducted two experiments based on general inference and $\alpha$-cut, and compared the experimental results. In the first experiment emotions were classified according to fuzzy inference. On the other hand in the second experiment emotions were classified according to $\alpha$-cut. Our experimental results showed that the classification of emotion based on a- cut was more definite that that based on fuzzy inference.

• 감성과학
• /
• 제15권2호
• /
• pp.283-296
• /
• 2012

본 연구는 SNS 사용자 간의 관계 유형을 감성공유관계와 정보공유관계로 분류하여 두 관계에 따른 사회감성모델을 구축하는 것이 목적이다. 이를 위해 먼저, 기존 문헌에 근거하여 92개의 감성을 수집하여 적합성 검증을 수행한 후 연구목적에 부합하는 26개의 사회감성을 추출하였다. 추출된 사회감성은 관계 유형에 따른 요인분석을 통해 12개의 대표 사회감성과 13개의 대표 사회감성을 도출하였다. 도출된 대표 사회감성은 다차원척도분석을 통하여 2차원의 공간에 매핑하여 사회감성 모델을 도출하였다. 도출된 사회감성 모델은 구조방정식모형 분석을 통해 통계적으로 유의하지 않은 요소들을 제거하였다. 타당성 검증 결과 적합도 지수를 통해 감성공유관계의 사회감성 모델(CFI:.887, TLI:.885, RMSEA:.094)과 정보공유관계의 사회감성 모델(CFI:.917, TLI:.900, RMSEA:.050)의 적합도가 나타났다. 본 연구의 결과로 사용자 관계 유형에 따라 상이한 사회감성 모델이 검증되었다. 본 연구를 통해 제시된 사회감성 모델은 SNS상에서 인간의 감성을 측정하는데 필요한 평가 자료로 활용될 수 있을 뿐만 아니라 향후 사회감성의 발전 방향을 제시할 수 있는 자료로 활용될 수 있을 것이라 사료된다.

• 대한인간공학회지
• /
• 제30권1호
• /
• pp.179-186
• /
• 2011

Objective: This study is to review three representative psychological perspectives that explain scientific construct of emotion, that are the discrete emotion model, appraisal model, and dimensional model. Background: To develop emotion sensitive interface is the fusion area of emotion and scientific technology, it is necessary to have a balanced mixture of both the scientific theory of emotion and practical engineering technology. Extensional theories of the emotional structure can provide engineers with relevant knowledge in functional application of the systems. Method: To achieve this purpose, firstly, literature review on the basic emotion model and the circuit model of discrete emotion model as well as representative theories was done. Secondly, review on the classical and modern theories of the appraisal model emphasizing cognitive appraisal in emotion provoking events was conducted. Lastly, a review on dimensional theories describing emotion by dimensions and representative theories was conducted. Results: The paper compared the three models based on the prime points of the each model. In addition, this paper also made a comment on a need for a comprehensive model an alternative to each model, which is componential model by Scherer(2001) describing numerous emotional aspects. Conclusion: However, this review suggests a need for an evolved comprehensive model taking consideration of social context effect and discrete neural circuit while pinpointing the limitation of componential model. Application: Insight obtained by extensive scientific research in human emotion can be valuable in development of emotion sensitive interface and emotion recognition technology.

• 감성과학
• /
• 제15권1호
• /
• pp.133-140
• /
• 2012

본 연구에서는 개인 맞춤형 감성인식이 가능한 실시간 개인화 프로세스를 개발하였다. 설계된 개인화 프로세스는 실시간 생체신호 중립밴드 알고리즘과 사용자의 주관적인 감성응답을 실시간으로 입력받아 개인 감성룰베이스를 업데이트하는 과정을 포함한다. 실시간 감성인식을 위해 PPG(Photoplethysmography), GSR(Galvanic skin reflex), SKT(Skin Temperature)를 자율신경계 신호를 측정하고 사용하였다. 개인화 프로세스를 적용한 감성인식과 적용하지 않은 감성인식의 일치도 평가는 대학생 10명(남 5명, 여 5명, 22.1세${\pm}$2.2세)의 피실험자를 대상으로 실시하였다. 45장의 이미지를 무작위로 제시하였으며, 5회 반복 측정하였다. 개인화 프로세스를 적용시켰을 때 약 71.67%의 일치도를 보였으며, 적용시키지 않았을 때보다 약 5배 이상 높은 일치도의 차이를 보였다. 본 연구에서는 개인화 프로세스가 실시간 개인 맞춤 감성인식에 매우 유용함을 보였다. 추후 개인화 프로세스는 다양한 감성 애플리케이션이나 서비스에서 감성인식 만족도를 높일 수 있을 것으로 판단된다.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2012년도 춘계학술대회 논문집
• /
• pp.881-882
• /
• 2012

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2013년도 춘계학술대회 논문집
• /
• pp.1041-1042
• /
• 2013

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2004년도 ICCAS
• /
• pp.1776-1780
• /
• 2004

In this paper, we propose the new emotion model and the robot behavior decision model based on proposed emotion model. As like in human, emotions are hierarchicalized in four levels (momentary emotions, mood, attitude, and personality) and are determined from the robot behavior and human responses. They are combined with motivation (which is determined from the external stimuli) to determine the robot behavior.