EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Brief speech samples reveal emotional states in daily life

Timo K Koch iD, Gabriella M Harari iD, Samuel D Gosling iD, Zachariah Marrero, Ramona Schoedel iD, Markus Bühner iD, Clemens Stachl iD

PNAS Nexus · 03.08.2026 · 5 (8)

DOI: 10.1093/pnasnexus/pgag263 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PNAS Nexus, online erschienen am 06.08.2026. Verfasst von Timo K Koch und 6 weiteren Autor:innen. Beteiligte Einrichtung: Institute of Behavioral Science and Technology, University of St Gallen, St Gallen 9000, Switzerland. Die Arbeit nennt 47 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Swiss National Science Foundation, National Science Foundation, German Academic Scholarship Foundation.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Does our speech—“what” we say and “how” we say it—reveal how we subjectively feel in daily life? Speech and emotion are often thought to be linked, but most supporting evidence comes from controlled laboratory settings and focuses on expressed, enacted, or externally labeled emotions, leaving open the question of whether naturalistic speech reflects subjective emotional experience in daily life. To address this question, we extracted both modern foundation-model embeddings and established speech-analysis variables from brief, prompted smartphone recordings, including Linguistic Inquiry and Word Count features and prosodic descriptors, and used them as inputs to supervised machine learning models predicting concurrent self-reported emotional states (934 participants; 12,285 observations). Cross-validated models evaluated on unseen participants captured information about self-reported emotional states (contentment: median Spearman ρ = 0.37; sadness: ρ = 0.24; arousal: ρ = 0.35), with spoken content represented by foundation-model text embeddings carrying the strongest emotional signal. Interpretability analyses provided further insights into the linguistic characteristics of everyday emotional language. These findings provide large-scale evidence that brief speech samples contain information about subjective emotional states in daily life.

Schlagwörter

Stichwörter der Autor:innen

emotional statesfoundation modelssmartphonesspeechvoice

Autor:innen und Institutionen

  • Institute of Behavioral Science and Technology, University of St Gallen, St Gallen 9000, Switzerland.
  • Department of Psychology, Ludwig-Maximilians-Universität München, Munich 80802, Germany.
  • Department of Communication, Stanford University, Stanford, CA 94305, USA.
  • Department of Psychology, The University of Texas at Austin, Austin, TX 78712, USA.
  • Department of Psychology, Charlotte Fresenius Hochschule, University of Psychology, Munich 80797, Germany.

ORCID-Profile: Timo K Koch, Gabriella M Harari, Samuel D Gosling, Ramona Schoedel, Markus Bühner, Clemens Stachl

Förderung

  • Swiss National Science Foundation
  • National Science Foundation
  • German Academic Scholarship Foundation

Zitieren

APA

Koch, T. K., Harari, G. M., Gosling, S. D., Marrero, Z., Schoedel, R., Bühner, M., & Stachl, C. (2026). Brief speech samples reveal emotional states in daily life. PNAS Nexus, 5(8). https://doi.org/10.1093/pnasnexus/pgag263

Vancouver

Koch TK, Harari GM, Gosling SD, Marrero Z, Schoedel R, Bühner M, et al. Brief speech samples reveal emotional states in daily life. PNAS Nexus. 2026;5(8). doi:10.1093/pnasnexus/pgag263

BibTeX
@article{Koch2026,
  author = {Koch, Timo K and Harari, Gabriella M and Gosling, Samuel D and Marrero, Zachariah and Schoedel, Ramona and Bühner, Markus and Stachl, Clemens},
  title = {Brief speech samples reveal emotional states in daily life},
  journal = {PNAS Nexus},
  year = {2026},
  volume = {5},
  number = {8},
  doi = {10.1093/pnasnexus/pgag263}
}
Kostenlos starten →