{"title":"Bioengineering system for research on human emotional response to external stimuli","authors":"Filatova Natalya Nikolaevna, Terekhin Sergey Alexeevich","doi":"10.1109/SIBIRCON.2015.7361841","DOIUrl":null,"url":null,"abstract":"Multichannel biotechnical system is designed to study human emotional reactions to external stimuli by recording and analyzing speech samples test or electroencephalogram signals. The system allows you to create complex stimuli for the perception that the subject uses a visual, acoustic or olfactory system. For registration of biological signals earmarked voice recording and electroencephalogram (EEG). Since stimulus material acts on three different channels of human perception, in the biotechnical system includes means of visual, audio-visual and olfactory stimulation. All system components are integrated into the control unit shown in the form of a personal computer with pre-installed special software. A feature of the system work scenario is the registration sample speech only after the marked changes in the electroencephalogram caused by the perception of the appropriate stimulus. This approach increases the credibility of the sampling of speech, which can then be used to form the rules of recognition of types of emotions. With this system were formed multimodal database comprising 280 different russian speakers phrases and corresponding EEG recording. The goals of experiments are studies attribute spaces are invariant to the physical nature of the signal and adequately reflects the emotional response of the subject, and testing the hypothesis of invariance of attributes allocated to the national characteristics of speech signals. In order to study the dynamics of emotional response developed a method that allows to explore a variety of discrete portions of biological signals. The technique is based on the idea of shifting the window, which is carried out static analysis. For each window are used as classical methods of calculation of the spectrum of the signal, and the methods of nonlinear dynamics. System showed enough high accuracy interpreting both the EEG pattern (83%) and with the speech samples (96 %) of the multimodal base emotions and objects taken from German base Emo-DB (92%). Evaluation of different emotional states a person can be used in the diagnosis of psychosomatic and psychomotor disturbances.","PeriodicalId":6503,"journal":{"name":"2015 International Conference on Biomedical Engineering and Computational Technologies (SIBIRCON)","volume":"24 1","pages":"13-17"},"PeriodicalIF":0.0000,"publicationDate":"2015-12-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2015 International Conference on Biomedical Engineering and Computational Technologies (SIBIRCON)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/SIBIRCON.2015.7361841","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1
Abstract
Multichannel biotechnical system is designed to study human emotional reactions to external stimuli by recording and analyzing speech samples test or electroencephalogram signals. The system allows you to create complex stimuli for the perception that the subject uses a visual, acoustic or olfactory system. For registration of biological signals earmarked voice recording and electroencephalogram (EEG). Since stimulus material acts on three different channels of human perception, in the biotechnical system includes means of visual, audio-visual and olfactory stimulation. All system components are integrated into the control unit shown in the form of a personal computer with pre-installed special software. A feature of the system work scenario is the registration sample speech only after the marked changes in the electroencephalogram caused by the perception of the appropriate stimulus. This approach increases the credibility of the sampling of speech, which can then be used to form the rules of recognition of types of emotions. With this system were formed multimodal database comprising 280 different russian speakers phrases and corresponding EEG recording. The goals of experiments are studies attribute spaces are invariant to the physical nature of the signal and adequately reflects the emotional response of the subject, and testing the hypothesis of invariance of attributes allocated to the national characteristics of speech signals. In order to study the dynamics of emotional response developed a method that allows to explore a variety of discrete portions of biological signals. The technique is based on the idea of shifting the window, which is carried out static analysis. For each window are used as classical methods of calculation of the spectrum of the signal, and the methods of nonlinear dynamics. System showed enough high accuracy interpreting both the EEG pattern (83%) and with the speech samples (96 %) of the multimodal base emotions and objects taken from German base Emo-DB (92%). Evaluation of different emotional states a person can be used in the diagnosis of psychosomatic and psychomotor disturbances.