Comparing accuracy in voice-based assessments of biological speaker traits across speech types - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Scientific Reports Année : 2023

Comparing accuracy in voice-based assessments of biological speaker traits across speech types

Piotr Sorokowski
  • Fonction : Auteur
  • PersonId : 1330433
Agata Groyecka-Bernard
  • Fonction : Auteur
Tomasz Frackowiak
  • Fonction : Auteur
Aleksander Kobylarek
  • Fonction : Auteur
Piotr Kupczyk
  • Fonction : Auteur
Agnieszka Sorokowska
  • Fonction : Auteur
Michał Misiak
  • Fonction : Auteur
Anna Oleszkiewicz
  • Fonction : Auteur
Katarzyna Bugaj
  • Fonction : Auteur
Małgorzata Włodarczyk
  • Fonction : Auteur

Résumé

Nonverbal acoustic parameters of the human voice provide cues to a vocaliser's sex, age, and body size that are relevant in human social and sexual communication, and also increasingly so for computer-based voice recognition and synthesis technologies. While studies have shown some capacity in human listeners to gauge these biological traits from unseen speakers, it remains unknown whether speech complexity improves accuracy. Here, in over 200 vocalisers and 1500 listeners of both sexes, we test whether voice-based assessments of sex, age, height and weight vary from isolated vowels and words, to sequences of vowels and words, to full sentences or paragraphs. We show that while listeners judge sex and especially age more accurately as speech complexity increases, accuracy remains high across speech types, even for a single vowel sound. In contrast, the actual heights and weights of vocalisers explain comparatively less variance in listener's assessments of body size, which do not vary systematically by speech type. Our results thus show that while more complex speech can improve listeners' biological assessments, the gain is ecologically small, as listeners already show an impressive capacity to gauge speaker traits from extremely short bouts of standardised speech, likely owing to within-speaker stability in underlying nonverbal vocal parameters such as voice pitch. We discuss the methodological, technological, and social implications of these results. Research on the biological and social communicative functions of nonverbal vocal parameters in the human voice has increased tremendously in the past two decades (for reviews see 1-3). Hundreds of studies have measured the nonverbal properties of the human voice and/or examined how these affect listeners' judgments in perception experiments. The overall aim is to understand what nonverbal features in the human voice can tell us about the speaker, from their emotional and motivational state or their physical traits such as their sex, age, and body size, to how such states and traits are perceived by others, as well as how vocal cues can offer insight into the communicative context of the vocal interaction 4,5. The knowledge gained from such research has far reaching theoretical implications, including uncovering the evolutionary origins of acoustic communication and its adaptive social functions 6. Indeed, vocal cues to sex, age and body size are important in the social and sexual communication systems of many animals 7 , including humans 8 , with ample evidence that selection has shaped the human voice to be highly sexually dimorphic and to signal, or exaggerate, body size 9,10. Research into the communicative functions of the human voice also has many practical applications, for example improving voice recognition technologies (for review see 11). However, the extent to which human listeners' voice-based judgments of biological or social traits reliably reflect the 'true' traits of vocalisers remains an open question, largely because findings vary across studies. It is not clear to what extent methodological differences play a role in this variation. Indeed, there remains surprisingly little consensus regarding the methods applied across voice studies, including those testing the exact same research question. Voice stimuli used in such research typically derive from audio recordings of vocalisers, but they vary widely in the type of speech recording content used for acoustic analysis and perception experiments, ranging from simple vowel sounds or nonverbal vocalisations to long trains of speech. This variation might

Domaines

Psychologie
Fichier principal
Vignette du fichier
Sorokowski:Pisanski et al 2023 - Voice-based perceptions of biological traits from different speech types.pdf (1.56 Mo) Télécharger le fichier
Origine Publication financée par une institution

Dates et versions

hal-04367399 , version 1 (30-12-2023)

Identifiants

Citer

Piotr Sorokowski, Agata Groyecka-Bernard, Tomasz Frackowiak, Aleksander Kobylarek, Piotr Kupczyk, et al.. Comparing accuracy in voice-based assessments of biological speaker traits across speech types. Scientific Reports, 2023, 13, ⟨10.1038/s41598-023-49596-y⟩. ⟨hal-04367399⟩

Collections

CNRS
4 Consultations
4 Téléchargements

Altmetric

Partager

Gmail Mastodon Facebook X LinkedIn More