Flexible voices: identity perception from variable vocal signals

Abstract

Human voices are extremely variable: The same person can sound very different depending on whether they are speaking, laughing, shouting or whispering. In order to successfully recognise someone from their voice, a listener needs to be able to generalise across these different vocal signals ('telling people together'). However, in most studies of voice identity processing to date, the substantial within-person variability has been eliminated through the use of highly controlled stimuli, thus focussing on how we tell people apart. We argue that this obscures our understanding of voice identity processing by controlling away an essential feature of vocal stimuli that may include diagnostic information. In this paper, we propose that we need to extend the focus of voice identity research to account for both 'telling people together' as well as 'telling people apart'. That is, we must account for whether, and to what extent, listeners can overcome within-person variability to obtain a stable percept of person identity from vocal cues. To do this, our theoretical and methodological frameworks need to be adjusted to explicitly include the study of within-person variability.

Working-memory disruption by task-irrelevant talkers depends on degree of talker familiarity

“…summarizes and illustrates these differences in the representation of familiar and unfamiliar voices, and how these differences may relate to working-memory disruption and distraction by concurrent speech. Consistent with recent advances in voice-identity research(Lavan et al, 2018a), we argue that a high degree of talker familiarity is needed to arrive at a stable representation of vocal identity. Moderate talker familiarity, however, creates uncertainty about vocal identity which, in turn, causes disruption of working memory and distraction by concurrent speech.…”

“…Family members and friends, by contrast, have heard the familiar talker's voice in a much wider range of contexts. Recent work (Lavan et al, 2018a;2018b) suggests that it is this variance in previous voice encounters that enables listeners to form a stable percept of person identity.…”

How many voices did you hear? Natural variability disrupts identity perception from unfamiliar voices

“…This approach has allowed us to gain insights into how we tell people apart via the distinguishing features of individual voices. It has, however, at the same time restricted our understanding of voice identity perception to this particular set of contexts, neglecting the study of the perceptual mechanisms that we use to compute stable and consistent representations of familiar voices despite the substantial within‐person variability (Lavan, Scott, & McGettigan, ; Lavan et al ., ). Similar issues have recently been highlighted for the face identity processing literature (Burton, ; Burton, Kramer, Ritchie, & Jenkins, ), opening up a fruitful new line of enquiry in this field.…”

“…Voices are highly variable. The same person can sound very different depending on the speaking context: For example, we modulate pitch, speech rate, and speaking style depending on whether we are giving a public lecture, talking to a friend, or singing (Kreiman, Park, Keating, & Alwan, ; Lavan, Burton, Scott, & McGettigan, ). Thus, within‐person variability is an inherent feature of the human voice that we encounter in all of our interactions.…”

A model of person perception from voices

“…Human voices are highly flexible (Lavan, Burton, Scott & McGettigan, 2019): Our voices can sound very different from one instance to another as a result of both volitional and involuntary modulations. For unfamiliar voice processing, this means that if listeners are given a number of variable samples of unfamiliar voices, their percepts of speaker characteristics may differ for each sample, regardless of the availability of information in the acoustic signal about the ground truth of the speakers' vocal anatomy.…”