Feasibility of Big Data Analytics to Assess Personality Based on Voice Analysis
Abstract
:1. Introduction
Voice Analysis for Paralinguistic Information Extraction
- Are there voice features related to the self-reported Big Five personality scores of individuals? In other words, are there specific voice features related to the externalization of individuals’ personality traits? It is expected, given the previous results analyzed, that personality is expressed by speech and voice markers, and these might be useful predictors of personality;
- Are voice features related to externalization the same as those related to the accuracy of personality judgments? Put another way, are voice features related to the close-others personality judgments and expert zero-acquaintance ratings the same as those that predict self-reported personality? It is expected that the expressive features would not be exactly the same as those used by others to make personality judgments;
- Can we design a feasible assessment setting that could be useful for an automatic personality assessment procedure? In other words, can we predict personality dimensions based on voice recordings obtained by ordinary devices in controlled, but not restricted, natural speech situations? It is expected that if voice features predict personality dimensions, these features can be processed and used in non-experimental conditions and, therefore, implemented in ordinary assessment contexts such as Asynchronous Video Interviews (AVIs).
2. Materials and Methods
2.1. Participants
2.2. Measures
2.2.1. Voice Features
2.2.2. Personality Assessment
2.3. Procedure
2.4. Data Analysis
3. Results
3.1. Correlation Analysis
3.1.1. Voice Features Related to Extraversion
3.1.2. Voice Features Related to Neuroticism
3.1.3. Voice Features Related to Openness
3.1.4. Voice Features Related to Agreeableness
3.1.5. Voice Features Related to Conscientiousness
3.1.6. Comparison with Results Obtained in a Different Set of Subjects
3.2. Prediction Analysis
4. Discussion
Supplementary Materials
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Acknowledgments
Conflicts of Interest
References
- Uleman, J.S.; Saribay, S.A. Initial impressions of others. In The Oxford Handbook of Personality and Social Psychology, 2nd ed.; Deaux, K., Snyder, M., Eds.; Oxford University Press: Oxford, UK, 2012; pp. 337–366. [Google Scholar]
- Funder, D.C.; Colvin, C.R. Friends and strangers: Acquaintanceship, agreement, and the accuracy of personality judgment. J. Personal. Soc. Psychol. 1988, 55, 149. [Google Scholar] [CrossRef] [PubMed]
- Ambady, N.; Rosenthal, R. Thin slices of expressive behavior as predictors of interpersonal consequences: A meta-analysis. Psychol. Bull. 1992, 111, 256. [Google Scholar] [CrossRef]
- Goffman, E. Gender Advertisements; Harper & Rowe: New York, NY, USA, 1979. [Google Scholar]
- Palese, T.; Schmid Mast, M. Interpersonal accuracy and interaction outcomes: Why and how reading others correctly has adaptive advantages in social interactions. In Social Intelligence: The Adaptive Advantages of Nonverbal Communication; Sternberg, R.J., Kostic’, A., Eds.; Palgrave-Macmillan: London, UK, 2019; pp. 305–331. [Google Scholar]
- Allport, G.W.; Cantril, H. Judging personality from voice. J. Soc. Psychol. 1934, 5, 37–55. [Google Scholar] [CrossRef]
- Polzehl, T. Personality in speech. In Assessment and Automatic Classification; Springer: Berlin/Heidelberg, Germany, 2015. [Google Scholar]
- Sherer, K.R. Personality markers in speech. In Social Markers in Speech; Scherer, K.R., Giles, H., Eds.; Cambridge University Press: Cambridge, MA, USA, 1978; pp. 147–209. [Google Scholar]
- Brunswick, E. Perception and the Representative Design of Experiments; University of California Press: Berkeley, CA, USA, 1956. [Google Scholar]
- Brown, B.; Bradshaw, J. Towards a social psychology of voice variations. In Recent Advances in Language Communication and Social Psychology; Giles, H., Clair, R.S., Eds.; Erlbaum: London, UK, 1985; pp. 144–181. [Google Scholar]
- Addington, D.W. The relationship of selected vocal characteristics to personality perception. Speech Monogr. 1968, 35, 492–503. [Google Scholar] [CrossRef]
- Aronovitch, C.D. The voice of personality: Stereotyped judgments and their relation to voice quality and sex of speaker. J. Soc. Psychol. 1976, 99, 207–220. [Google Scholar] [CrossRef]
- Furham, A. Language and personality. In Handbook of Language and Social Psychology; Giles, H., Robinson, W.P., Eds.; John Wiley: Chichester, UK, 1990. [Google Scholar]
- Mairesse, F.; Walker, M.A.; Mehl, M.R.; Moore, R.K. Using linguistic cues for the automatic recognition of personality in conversation and text. J. Artif. Intell. Res. 2007, 30, 457–500. [Google Scholar] [CrossRef]
- Hu, C.; Wang, Q.; Short, L.A.; Fu, G. Speech spectrum’s correlation with speakers’ Eysenck Personality Traits. PLoS ONE 2012, 7, e33906. [Google Scholar] [CrossRef]
- Belin, P.; Boehme, B.; McAleer, P. The sound of trustworthiness: Acoustic-based modulation of perceived voice personality. PLoS ONE 2017, 12, e0185651. [Google Scholar] [CrossRef]
- McAleer, P.; Todorov, A.; Belin, P. How do you say ‘Hello’? Personality impressions from brief novel voices. PLoS ONE 2014, 9, e90779. [Google Scholar] [CrossRef]
- Metze, F.; Black, A.; Polzehl, T. A review of personality in voice-based man machine interaction. In Human-Computer Interaction. Interaction Techniques and Environments—14th International Conference, HCI International 2011, Orlando, FL, USA, 9–14 July 2011; Springer: Berlin/Heidelberg, Germany, 2011; pp. 358–367. [Google Scholar]
- Polzehl, T.; Möller, S.; Metze, F. Automatically assessing personality from speech. In Proceedings of the International Conference on Semantic Computing (ICSC 2010), Pittsburgh, PA, USA, 22–24 September 2010; pp. 1–6. [Google Scholar]
- Stern, J.; Schild, C.; Jones, B.C.; DeBruine, L.M.; Hahn, A.; Puts, D.A.; Arslan, R.C. Do voices carry valid information about a speaker’s personality? J. Res. Personal. 2021, 92, 104092. [Google Scholar] [CrossRef]
- Marrero, Z.N.; Gosling, S.D.; Pennebaker, J.W.; Harari, G.M. Evaluating voice samples as a potential source of information about personality. Acta Psychol. 2022, 230, 103740. [Google Scholar] [CrossRef] [PubMed]
- Schmitt, A.; Zierau, N.; Janson, A.; Leimeister, J.M. Voice as a contemporary frontier of interaction design. In Proceedings of the European Conference on Information Systems (ECIS), Virtual, 14–16 June 2021. [Google Scholar]
- Van Zant, A.B.; Berger, J. How the voice persuades. J. Personal. Soc. Psychol. 2020, 118, 661. [Google Scholar] [CrossRef] [PubMed]
- Breil, S.M.; Osterholz, S.; Nestler, S.; Back, M.D. 13 contributions of nonverbal cues to the accurate judgment of personality traits. In The Oxford Handbook of Accurate Personality Judgment; Oxford University Press: Oxford, UK, 2021; pp. 195–218. [Google Scholar]
- Gocsál, Á. Female listeners’ personality attributions to male speakers: The role of acoustic parameters of speech. Pollack Period. 2009, 4, 155–165. [Google Scholar] [CrossRef]
- Aylett, M.P.; Vinciarelli, A.; Wester, M. Speech synthesis for the generation of artificial personality. IEEE Trans. Affect. Comput. 2017, 11, 361–372. [Google Scholar] [CrossRef]
- Cannata, D.; Breil, S.M.; Lepri, B.; Back, M.D.; O’Hora, D. Toward an integrative approach to nonverbal personality detection: Connecting psychological and artificial intelligence research. Technol. Mind Behav. 2022, 3, 1–16. [Google Scholar] [CrossRef]
- Martin, A.F.; Przybocki, M.A. The NIST speaker recognition evaluations: 1996–2001. In Proceedings of the Speaker Recognition Workshop, Crete, Greece, 18–22 June 2001. [Google Scholar]
- Alvin, M.P.; Martin, A. NIST speaker recognition evaluation chronicles. In Proceedings of the Odyssey 2004, The Speaker and Language Recognition Workshop, Toledo, Spain, 1–3 June 2004. [Google Scholar]
- Przybocki, M.A.; Martin, A.F.; Le, A.N. NIST speaker recognition evaluations utilizing the Mixer corpora—2004, 2005, 2006. IEEE Trans. Audio Speech Lang. Process. 2007, 15, 1951–1959. [Google Scholar] [CrossRef]
- Satt, A.; Sorin, A.; Toledo-Ronen, O.; Barkan, O.; Kompatsiaris, I.; Kokonozi, A.; Tsolaki, M. Evaluation of speech-based protocol for detection of early-stage dementia. In Proceedings of the Interspeech 2013, Lyon, France, 25–29 August 2013; pp. 1692–1696. [Google Scholar]
- Satt, A.; Hoory, R.; König, A.; Aalten, P.; Robert, P.H. Speech-based automatic and robust detection of very early dementia. In Proceedings of the Fifteenth Annual Conference of the International Speech Communication Association, Singapore, 14–18 September 2014. [Google Scholar]
- Weiner, J.; Herff, C.; Schultz, T. Speech-Based Detection of Alzheimer’s Disease in Conversational German. In Proceedings of the Interspeech 2016, San Francisco, CA, USA, 8–12 September 2016; pp. 1938–1942. [Google Scholar]
- Skodda, S.; Schlegel, U. Speech rate and rhythm in Parkinson’s disease. Mov. Disord. Off. J. Mov. Disord. Soc. 2008, 23, 985–992. [Google Scholar] [CrossRef]
- Rusz, J.; Cmejla, R.; Ruzickova, H.; Ruzicka, E. Quantitative acoustic measurements for characterization of speech and voice disorders in early untreated Parkinson’s disease. J. Acoust. Soc. Am. 2011, 129, 350–367. [Google Scholar] [CrossRef]
- Baykaner, K.R.; Huckvale, M.; Whiteley, I.; Andreeva, S.; Ryumin, O. Predicting fatigue and psychophysiological test performance from speech for safety-critical environments. Front. Bioeng. Biotechnol. 2015, 3, 124. [Google Scholar] [CrossRef]
- Hall, J.A.; Gunnery, S.D.; Letzring, T.D.; Carney, D.R.; Colvin, C.R. Accuracy of judging affect and accuracy of judging personality: How and when are they related? J. Personal. 2017, 85, 583–592. [Google Scholar] [CrossRef]
- Schuller, B.; Valstar, M.; Eyben, F.; McKeown, G.; Cowie, R.; Pantic, M. Avec 2011–the first international audio/visual emotion challenge. In International Conference on Affective Computing and Intelligent Interaction; Springer: Berlin, Germany, 2011; pp. 415–424. [Google Scholar]
- Eyben, F.; Wöllmer, M.; Schuller, B. Opensmile: The munich versatile and fast open-source audio feature extractor. In Proceedings of the 18th ACM International Conference on Multimedia, Firenze, Italy, 29 October 2010; ACM: New York, NY, USA, 2010; pp. 1459–1462. [Google Scholar]
- Schuller, B.; Steidl, S.; Batliner, A.; Nöth, E.; Vinciarelli, A.; Burkhardt, F.; van Son, R.; Weninger, F.; Eyben, F.; Bocklet, T.; et al. A survey on perceived speaker traits: Personality, likability, pathology, and the first challenge. Comput. Speech Lang. 2015, 29, 100–131. [Google Scholar] [CrossRef]
- Mohammadi, G.; Vinciarelli, A. Automatic personality perception: Prediction of trait attribution based on prosodic features. IEEE Trans. Affect. Comput. 2012, 3, 273–284. [Google Scholar] [CrossRef]
- Ponce-López, V.; Chen, B.; Oliu, M.; Corneanu, C.; Clapés, A.; Guyon, I.; Baró, X.; Escalante, H.J.; Escalera, S. ChaLearn LAP 2016: First Round Challenge on First Impressions—Dataset and Results. In Computer Vision—ECCV 2016 Workshops, ECCV 2016; Hua, G., Jégou, H., Eds.; Lecture Notes in Computer Science; Springer: Berlin/Heidelberg, Germany, 2016; Volume 9915. [Google Scholar]
- Palmero, C.; Junior, J.C.J.; Clapés, A.; Guyon, I.; Tu, W.W.; Moeslund, T.B.; Escalera, S. Understanding Social Behavior in Dyadic and Small Group Interactions: Preface. PMLR 2022, 173, 1–3. [Google Scholar]
- Escalante, H.J.; Kaya, H.; Salah, A.A.; Escalera, S.; Güçlütürk, Y.; Güçlü, U.; Van Lier, R. Modeling, recognizing, and explaining apparent personality from videos. IEEE Trans. Affect. Comput. 2020, 13, 894–911. [Google Scholar] [CrossRef]
- Palmero, C.; Barquero, G.; Junior, J.C.J.; Clapés, A.; Núnez, J.; Curto, D.; Escalera, S. Chalearn LAP challenges on self-reported personality recognition and non-verbal behavior forecasting during social dyadic interactions: Dataset, design, and results. PMLR 2022, 173, 4–52. [Google Scholar]
- Palmero, C.; Selva, J.; Smeureanu, S.; Junior, J.; Jacques, C.S.; Clapés, A.; Escalera, S. Context-aware personality inference in dyadic scenarios: Introducing the udiva dataset. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, Virtual, 5–9 January 2021; pp. 1–12. [Google Scholar]
- Ryumina, E.; Ryumin, D.; Markitantov, M.; Kaya, H.; Karpov, A. Multimodal personality traits assessment (MuPTA) corpus: The impact of spontaneous and read speech. In Proceedings of the ISCA International Conference INTERSPEECH, Dublin, Ireland, 20–24 August 2023; pp. 4049–4053. [Google Scholar]
- Ryumina, E.; Markitantov, M.; Ryumin, D.; Karpov, A. OCEAN-AI framework with EmoFormer cross-hemiface attention approach for personality traits assessment. Expert Syst. Appl. 2024, 239, 122441. [Google Scholar] [CrossRef]
- Koutsoumpis, A.; Ghassemi, S.; Oostrom, J.K.; Holtrop, D.; Van Breda, W.; Zhang, T.; de Vries, R.E. Beyond traditional interviews: Psychometric analysis of asynchronous video interviews for personality and interview performance evaluation using machine learning. Comput. Hum. Behav. 2024, 154, 108128. [Google Scholar] [CrossRef]
- Hickman, L.; Bosch, N.; Ng, V.; Saef, R.; Tay, L.; Woo, S.E. Automated video interview personality assessments: Reliability, validity, and generalizability investigations. J. Appl. Psychol. 2022, 107, 1323. [Google Scholar] [CrossRef]
- Allport, G.W. Personality: A Psychological Interpretation; Holt: New York, NY, USA, 1937. [Google Scholar]
- Cordero, A.; Pamós, A.; Seisdedos, N. Inventario de Personalidad NEO Revisado (NEO-PI-R)—Inventario NEO Reducido de Cinco Factores (NEO-FFI). In Manual Profesional. Adaptación Española, 3rd ed.; Tea Ediciones: Madrid, Spain, 2008. [Google Scholar]
- Breiman, L. Random forests. Mach. Learn. 2001, 45, 5–32. [Google Scholar] [CrossRef]
- Pedregosa, F.; Varoquaux, G.; Gramfort, A.; Michel, V.; Thirion, B.; Grisel, O.; Vanderplas, J. Scikit-learn: Machine learning in Python. J. Mach. Learn. Res. 2011, 12, 2825–2830. [Google Scholar]
- Jensen, M. Personality traits and nonverbal communication patterns. Int. J. Soc. Sci. Stud. 2016, 4, 57. [Google Scholar] [CrossRef]
- Gavrilescu, M. Study on determining the Big-Five personality traits of an individual based on facial expressions. In Proceedings of the 2015 E-Health and Bioengineering Conference (EHB), Iasi, Romania, 19–21 November 2015; pp. 1–6. [Google Scholar]
- Youyou, W.; Kosinski, M.; Stillwell, D. Computer-based personality judgments are more accurate than those made by humans. Proc. Natl. Acad. Sci. USA 2015, 112, 1036–1040. [Google Scholar] [CrossRef] [PubMed]
- Argamon, S.; Dhawle, S.; Koppel, M.; Pennebaker, J.W. Lexical predictors of personality type. In Proceedings of the 2005 Joint Annual Meeting of the Interface and the Classification Society of North America, St. Louis, MO, USA, 8–12 June 2005; pp. 1–16. [Google Scholar]
- Pennebaker, J.W.; King, L.A. Linguistic styles: Language use as an individual difference. J. Personal. Soc. Psychol. 1999, 77, 1296–1312. [Google Scholar] [CrossRef] [PubMed]
- Hosoda, M.; Stone-Romero, E.F.; Coats, G. The effects of physical attractiveness on job-related outcomes: A meta-analysis of experimental studies. Pers. Psychol. 2003, 56, 431–462. [Google Scholar] [CrossRef]
- Guidi, A.; Gentili, C.; Scilingo, E.P.; Vanello, N. Analysis of speech features and personality traits. Biomed. Signal Process. Control 2019, 51, 1–7. [Google Scholar] [CrossRef]
- Delgado-Gómez, D.; Masó-Besga, A.E.; Aguado, D.; Rubio, V.J.; Sujar, A.; Bayona, S. Automatic personality assessment through movement analysis. Sensors 2022, 22, 3949. [Google Scholar] [CrossRef]
Low-Level Descriptor | Description and Interpretation |
---|---|
Loudness | It is a measure of subjective perception of sound pressure. |
Zero crossing rate | It is the number of times the amplitude crosses the zero value in a given interval. |
Psychoacoustic sharpness | It is how much a sound’s spectrum is in the high end. |
Harmonicity | It represents the degree of acoustic periodicity. An HNR of 0 dB means that there is equal energy in the harmonics and in the noise. |
MFCC 1–10 | Mel Frequency Cepstral Coefficients are a representation of the short-term power spectrum of a signal in a psychoacoustic (Mel) scale. |
Kurtosis | It is a statistical measure that is used to describe the distribution of a signal, it measures extreme values in tail relative to a normal distribution. High kurtosis has heavy tails or outliers and low kurtosis has light tails or lack of outliers. |
Skewness | It is a statistical measure that is used to describe symmetry. Skewness near zero means symmetric data. |
Jitter and Shimmer | Both represent the variations in vibration of the vocal chords. Jitter is the variability in frequency and shimmer is the variability in amplitude. |
Spectral Flux | It is a measure of how quickly the power spectrum of a signal is changing. |
Voicing | Vibration of the vocal chords. |
Functional | Description |
---|---|
Statistical functionals (23) | Arithmetic mean (1), root quadratic mean (2), standard deviation (3), flatness (4), skewness (5), kurtosis (6), quartiles (7–9), inter-quartile ranges (10–12), 1% percentile (13), 99% percentile (14), percentile range 1–99% (15), percentage of frames where the contour is above the minimum + 25% (16), 50% (17) and 90% (18) of the range, percentage of frames where contour is rising (19), maximum (20), mean (21), minimum (22) and standard deviation of segment length (23). |
Regression functionals (4) | Linear regression slope (1) and corresponding error (2), quadratic regression coefficient (3) and error (4). |
Local minima/maxima related functionals (9) | Mean and standard deviation of rising and falling slopes (1–4), mean (5) and standard deviation (6) of inter maxima distances, amplitude mean of maxima (7), amplitude range of maxima (8) and minima (9). |
Others (6) | Linear Prediction Coefficients (LPC) (1 to 5), Linear Prediction (LP) gain (6) |
M | SD | Alpha | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | ||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
1 | Self-N | 32.89 | 7.65 | 0.83 | −0.25 * (0.011) | −0.17 (0.076) | −0.24 * (0.017) | −0.27 ** (0.005) | 0.32 ** (0.001) | −0.13 (0.200) | −0.02 (0.828) | −0.05 (0.583) | −0.06 (0.537) | 0.19 (0.057) | −0.14 (0.168) | −0.09 (0.367) | 0.10 (0.316) | −0.19 (0.060) |
2 | Self-E | 45.01 | 7.32 | 0.87 | 0.22 * (0.024) | 0.15 (0.134) | 0.10 (0.332) | 0.11 (0.289) | 0.51 ** (<0.001) | 0.03 (0.789) | −0.07 (0.477) | −0.09 (0.373) | −0.14 (0.170) | 0.31 ** (0.001) | 0.10 (0.297) | 0.03 (0.770) | 0.05 (0.581) | |
3 | Self-O | 43.52 | 8.32 | 0.87 | 0.09 (0.354) | 0.11 (0.276) | 0.08 (0.419) | −0.07 (0.459) | 0.13 (0.176) | 0.09 (0.370) | 0.013 (0.899) | −0.15 (0.142) | −0.10 (0.312) | 0.14 (0.150) | −0.07 (0.451) | 0.19 (0.090) | ||
4 | Self-A | 40.96 | 6.69 | 0.77 | 0.37 ** (<0.001) | 0.11 (0.258) | −0.02 (0.835) | 0.01 (0.892) | 0.36 ** (<0.001) | −0.05 (0.594) | 0.13 (0.192) | −0.15 (0.130) | 0.07 (0.461) | 0.14 (0.150) | 0.27 * (0.022) | |||
5 | Self-C | 43.57 | 8.02 | 0.88 | 0.01 (0.959) | 0.08 (0.422) | 0.03 (0.730) | 0.00 (0.966) | 0.41 ** (<0.001) | 0.00 (0.966) | −0.08 (0.396) | 0.01 (0.929) | 0.21 * (0.035) | 0.34 ** (<0.001) | ||||
6 | Rel-N | 1.92 | 0.67 | S.I. | −0.07 (0.467) | −0.13 (0.210) | −0.05 (0.601) | 0.02 (0.809) | 0.17 (0.093) | −0.06 (0.537) | 0.07 (0.170) | 0.16 (0.278) | 0.04 (0.097) | |||||
7 | Rel-E | 2.32 | 0.69 | S.I. | 0.15 (0.127) | −0.04 (0.665) | −0.08 (0.411) | −0.06 (0.524) | 0.27 ** (0.007) | 0.13 (0.188) | −0.13 (0.217) | −0.18 (0.066) | ||||||
8 | Rel-O | 2.30 | 0.66 | S.I. | 0.16 (0.120) | 0.04 (0.711) | 0.07 (0.464) | 0.03 (0.753) | 0.16 (0.101) | 0.08 (0.407) | −0.12 (0.218) | |||||||
9 | Rel-A | 2.04 | 0.76 | S.I. | 0.06 (0.520) | 0.15 (0.121) | 0.16 (0.119) | 0.15 (0.138) | 0.04 (0.683) | 0.25 * (0.010) | ||||||||
10 | Rel-C | 2.50 | 0.61 | S.I. | 0.04 (0.675) | 0.07 (0.492) | 0.00 (0.985) | 0.17 (0.086) | 0.27 ** (0.006) | |||||||||
11 | Exp-N | 8.53 | 3.46 | S.I. | −0.38 ** (<0.001) | −0.23 * (0.017) | 0.35 ** (<0.001) | −0.21 * (0.032) | ||||||||||
12 | Exp-E | 7.80 | 3.71 | S.I. | 0.48 ** (<0.001) | −0.07 (0.464) | −0.10 (0.298) | |||||||||||
13 | Exp-O | 6.91 | 3.79 | S.I. | −0.13 (0.204) | 0.04 (0.715) | ||||||||||||
14 | Exp-A | 8.37 | 3.17 | S.I. | 0.10 (0.314) | |||||||||||||
15 | Exp-C | 8.22 | 3.52 | S.I. |
Self (Sample 1) | Close Others (Sample 1) | Experts (Sample 1) | ||||
---|---|---|---|---|---|---|
Trait | Feature | Correlation (p-Value) | Feature | Correlation (p-Value) | Feature | Correlation (p-Value) |
Neuroticism | voicingFinalUnclipped_sma_de_percentile1.0 | 0.37 *** (<0.001) | mfcc_sma[3]_minRangeRel | 0.32 ** (0.004) | pcm_Mag_spectralRollOff75.0_sma_quartile1 | −0.47 *** (<0.001) |
Extraversion | mfcc_sma[3]_lpgain | 0.47 *** (<0.001) | voicingFinalUnclipped_sma_de_iqr1-3 | 0.40 *** (<0.001) | audspec_lengthL1norm_sma_iqr2-3 | 0.48 *** (<0.001) |
Open to Experience | audspec_lengthL1norm_sma_de_upleveltime25 | 0.36 ** (0.0013) | F0final_sma_de_upleveltime50 | −0.41 *** (<0.001) | mfcc_sma[10]_lpc3 | −0.36 ** (0.0011) |
Agreeableness | audspec_lengthL1norm_sma_de_meanSegLen | −0.33 ** (0.004) | mfcc_sma[5]_risetime | −0.41 *** (<0.001) | pcm_Mag_spectralRollOff75.0_sma_linregc1 | 0.35 ** (0.002) |
Conscientiousness | pcm_Mag_spectralVariance_sma_de_skewness | 0.30 ** (0.008) | pcm_Mag_psySharpness_sma_lpc2 | −0.30 ** (0.007) | mfcc_sma[5]_lpc4 | 0.44 *** (<0.001) |
Trait | Summary |
---|---|
Neuroticism | Neuroticism voice features that seem to be related to either instability in voicing or the distribution of energy more biased toward lower frequencies, which could be related to lower volume speech production. Neuroticism is associated with parameters, indicating that most energy is concentrated in low energy levels and could be related to very low voice volume. Additionally, in people with high levels of neuroticism, the smallest variations in the probability of voicing tend to be relatively large, which could indicate instabilities in the production of voiced speech. |
Extraversion | Extraversion is related to voice features that seem to be related to longer and more intense pronunciation of vowels and high-energy phones and also to larger variations in voice volume and speaking rate. Features related to Extraversion measure the predictability of the energy in different frequency bands based on recent past values, which is higher when high-energy phonemes, such as vowels, are prolonged and exhibit high energy. Additionally, greater variation in loudness appears to indicate that a larger range of variations in voice volume correlates with higher levels of Extraversion. |
Open to Experience | Open to Experience people are related to voice features that are related to either particular intonations (faster increases and slower decreases in pitch) or to having significant variations in voice volume for longer periods of time. |
Agreeableness | Agreeableness is related to voice features such as the raising slopes of the spectral roll-off, which implies rapid increases in high frequency energy in the voice. In general, voice features related to Agreeableness seem to be related with changes in the voice volume or frequency distribution, perhaps representing a more variable voice. |
Conscientiousness | Conscientiousness presents higher correlation with voice features that are difficult to interpret. It is also the personality dimension with the lowest maximum correlations (with the exception of the correlation with the expert-provided personality scores). |
All Cases | Low Trait Cases | Medium Trait Cases | High Trait Cases | |
---|---|---|---|---|
Self-asessment | ||||
Neuroticism | 50 | 34.5 | 61.2 | 44.0 |
Extraversion | 46 | 37.0 | 62.7 | 18.2 |
Open to Experience | 46 | 35.7 | 65.3 | 17.4 |
Agreeableness | 53 | 46.1 | 72.5 | 17.4 |
Conscientiousness | 43 | 23.1 | 72.0 | 4.1 |
Expert-assessment | ||||
Neuroticism | 57 | 42.8 | 81.1 | 10.5 |
Extraversion | 60 | 48.3 | 76.4 | 25 |
Open to Experience | 57 | 32 | 79.2 | 31.2 |
Agreeableness | 56 | 43.7 | 78.4 | 11.8 |
Conscientiousness | 53 | 57.7 | 70.0 | 12.5 |
Close Others-assessment | ||||
Neuroticism | 54 | 29.6 | 76.4 | 22.2 |
Extraversion | 57 | 15.4 | 65.1 | 61.4 |
Open to Experience | 45 | 18.2 | 47.9 | 48.8 |
Agreeableness | 52 | 53.8 | 52.4 | 50.0 |
Conscientiousness | 54 | 0 | 47.4 | 64.3 |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Rubio, V.J.; Aguado, D.; Toledano, D.T.; Fernández-Gallego, M.P. Feasibility of Big Data Analytics to Assess Personality Based on Voice Analysis. Sensors 2024, 24, 7151. https://doi.org/10.3390/s24227151
Rubio VJ, Aguado D, Toledano DT, Fernández-Gallego MP. Feasibility of Big Data Analytics to Assess Personality Based on Voice Analysis. Sensors. 2024; 24(22):7151. https://doi.org/10.3390/s24227151
Chicago/Turabian StyleRubio, Víctor J., David Aguado, Doroteo T. Toledano, and María Pilar Fernández-Gallego. 2024. "Feasibility of Big Data Analytics to Assess Personality Based on Voice Analysis" Sensors 24, no. 22: 7151. https://doi.org/10.3390/s24227151
APA StyleRubio, V. J., Aguado, D., Toledano, D. T., & Fernández-Gallego, M. P. (2024). Feasibility of Big Data Analytics to Assess Personality Based on Voice Analysis. Sensors, 24(22), 7151. https://doi.org/10.3390/s24227151