350 rub
Journal Biomedical Radioelectronics №12 for 2013 г.
Article in number:
Acoustic characteristics of speech signal and psychophysiological correlates of different emotional states in the model experiments
Keywords:
speech signal
nonlinear analysis
emotional state
gender differences
EEG
variation cardio measure
psychological testing
Authors:
N. N. Lebedeva - Dr.Sc.(Biol.), Head of Laboratory, Institute of Higher Nervous Activity and Neurophysiology of RAS (Moscow)
R. K. Potapova - Dr.Sc. (Phylol.), Professor, Head of Department of Applied and Experimental Linguistics, Moscow State Linguistic University
E. D. Karimova - Junior Research Scientist, Institute of Higher Nervous Activity and Neurophysiology of RAS
A.V. Vekhov - Junior Research Scientist, Institute of Higher Nervous Activity and Neurophysiology of RAS
E. A. Kazimirova - Student, Moscow State University n.a. M.V. Lomonosova
R. K. Potapova - Dr.Sc. (Phylol.), Professor, Head of Department of Applied and Experimental Linguistics, Moscow State Linguistic University
E. D. Karimova - Junior Research Scientist, Institute of Higher Nervous Activity and Neurophysiology of RAS
A.V. Vekhov - Junior Research Scientist, Institute of Higher Nervous Activity and Neurophysiology of RAS
E. A. Kazimirova - Student, Moscow State University n.a. M.V. Lomonosova
Abstract:
The main aim of our work was studying the non-linear characteristic of voice signal in perspective of verification of people emotional condition and an impact, individual psychophysiological features have on this characteristic. It was studied in model experiment, during which each participant should plunge in each emotional states and then read four sentences. We recruited twenty two people in experiment - twenty non-actor (10 men, 10 women) and two professional actors (man and woman). There were five states: neutrality, happiness, anger, sorrow and fear. We obtained the non-linear characteristic D2 of voice signal. In order to identify individual psychophysiological profile for each participant, ECG-recording, psychological tests (including Eysenck-s test) and EEG-recording were used.
Non-actors and actors results comparing have shown, that typical D2 dynamic is decrease during happiness and anger for all participants; but in sorrow and fear states D2 slightly increase for women and extroverts and decrease for men and introverts. These trends for Professional actors were more remarkable but non-actors also demonstrated the same. So, dynamic for D2 of women is similar to extroverts and of men is similar to introverts.
In addition to measuring the level of introversions and extraversions, the level of neuroticism were measured as well. It has been shown, that people with high level of this factor demonstrated higher meanings of D2. It must be considered, that the majority part of this group was formed by women.
After the division of all participants into several groups according their alpha-rhythm capacity we found out that D2 of people with ulterior rhythm changed hardly noticeable between different emotional states. For groups with alpha-rhythm varying degrees of capacity, value of D2 depends on the level of neuroticism. Besides the level of neuroticism influenced on the ratio of D2 values between different emotional states.
In addition, five experts were offered all audio files for marking. According to the expert estimate, the most successful modeling was demonstrated by people with high level of extraversion and neuroticism at once. But for all that, experts could not make out emotion, which had right dynamic of D2.
We may conclude that use non-linear voice characteristic may be more successful than expert estimation, at least for emotion directed outward (happiness, anger).
Pages: 35-42
References
- Morozov V.P. Iskusstvo i nauka obshheniya: neverbal'naya kommunikacziya. M.: IP RAN. 1998. 160 s.
- Sedlacek K., Sychra A. Hudba a slovo z experimcntelniho hlediska. Praga. 1962.
- Ostwald P. Soundmaking. Springield. 1963
- Popov V.A., Simonov P.V., Tishhenko A.G., Frolov M.V., Xachatur'yancz L.S. Analiz intonaczionnoj xarakteristiki rechi kak pokazatelya e'moczional'nogo sostoyaniya cheloveka v usloviyax kosmicheskogo naleta // Zhurnal vy'sshej nervnoj deyatel'nosti. 1966. T. 16. Vy'p. 6. C. 974-983.
- Luk'yanov A.N., Frolov M.V. Signaly' sostoyaniya cheloveka-operatora. M.: Nauka. 1969.
- Frolov M.V. Differenczirovanie stepeni i znaka e'moczional'nogo napryazheniya u cheloveka po izmeneniyu intonaczionny'x xarakteristik rechi // V sb.: «Fiziologicheskie osobennosti polozhitel'ny'x i otriczatel'ny'x e'moczional'ny'x sostoyanij». M.: Nauka. 1972. S. 128-132.
- Simonov P. V., Frolov M. V., Taubkin V. L. Invariantny'j metod raspoznaniya e'moczional'nogo sostoyaniya gruppy' diktorov po ix rechi // Zhurnal vy'sshej nervnoj deyatel'nosti. 1976. T. 26. № I.
- Frolov M.V. Taubkin V.L. O vliyanii e'moczional'nogo sostoyaniya diktora na nekotory'e parametry' rechevogo signala // V sb. «Rech' i e'moczii». L. 1975.
- Frolov M.V., Taubkin V.L., Luzhbin N.A. Algoritm raspoznavaniya e'moczional'ny'x sostoyanij po realizacziyam glavnoudarny'x glasny'x zvukov // V sb.: «Rech', e'moczii i lichnost'». L.: Nauka. 1978. S. 190-194.
- Bachorouski Jo-Anne. Vocal expression and perception of emotion // Current Directions in Psychological Science. 1999. V. 9. № 2. P.53-57.
- Rothkrantz L.J.M., Wiggers P., van Wees J.W.A., van Vark R.J. Voice Stress Analysis. Proceedings of Text. Speech and Dialogues. 2004.
- Petri Laukka. Vocal Expression of Emotion. Discrete-emotions and Dimensional Accounts. Acta Universitatis Upsaliensis // Comprehensive Summaries of Uppsala Dissertations from the Faculty of Social Sciences. 2004. V.141. P.80.
- Mimmi Forsell. Acoustical Correlates of Perceived Emotions in Speech. Master of Science Thesis. Stockholm. Sweden. 2007.
- Schuller B. Voice and Speech Analysis in Search of States and Traits. In: Salah A.A., Gevers T. (eds.) Computer Analisis of Human Behavior. Springer. Heidelberg. 2011. P. 227-253.
- Bachorouski Jo-Anne. Vocal expression and perception of emotion // Current Directions in Psychological Science. 1999. V. 9. № 2. P. 53-57.
- Frijda N. Recognition of emotion // Advances in Exp. Soc. Psychol. 1969. V. 4. R. 167-223.
- SÃO LUÍS CASTRO, CÉSAR F. LIMA. Recognizing emotions in spoken language: A validated set of Portuguese sentences and pseudosentences for research on emotional prosody // Behavior Research Methods. 2010. V. 42 (1). P. 74-81.
- Kim J. Bimodal emotion recognition using speech and physiological changes. In M. Grimm, K. Kroschel (Ed.). Robust Speech Rcognition and Understanding. P. 265-280.
- Bachorouski Jo-Anne. Vocal expression and perception of emotion // Current Directions in Psychological Science. 1999. V. 9. № 2. P.53-57.
- Pashina A. X., Morozov V. P. Opoznavanie lichnosti po golosu na osnove ego normal'nogo i invertirovannogo vo vremeni zvuchaniya // Psixologicheskij zhurnal. 1990. T. 11. № 3. S. 70-78.
- Lebedeva N.N., Karimova E.D., Kazimirova E.A. Akusticheskie xarakteristiki rechi i individual'ny'e osobennosti cheloveka // Biomediczinskaya radioe'lektronika. 2013. № 6. S. 29-35.
- Pashina A.X. K probleme raspoznavaniya e'moczional'nogo konteksta zvukovoj rechi // Voprosy' psixologii. 1991. № 1. S. 88-96.
- Lebedeva N.N., Sidorova O.A., Maragej R.A., Kotrovskaya A.N. Fraktal'ny'j analiz rechevogo signala pri razlichenii e'moczional'nogo sostoyaniya v model'ny'x e'ksperimentax // Biomediczinskaya radioe'lektronika. 2010. № 1. S. 3-7.
- Filatova N.N., Sidorov K.V. Model' interpretaczii znaka e'moczij po estestvennoj rechi // Izvestiya JuFU, 2012. № 9.