Communicative Characteristics of Voices Perceived as Feminine, Masculine, and Neutral: A Study with Brazilian Listeners


Objective: To analyze communicative characteristics of voices perceived as feminine, masculine, and neutral by Brazilian listeners.
Methods: Recordings from 11 cis women, 11 trans women, 7 trans men, 11 cis men, and 7 non-binary individuals were evaluated for gender perception in voice (100-point visual analog scale ranging from very masculine to very feminine) by 101 cisgender participants. The categorization of voices as masculine, feminine, or neutral was based on applying a z-score to the auditory-perceptual judgment (neutral point (0) ± 1 standard deviation). The neutrality range was established with an upper limit of 27.6 and a lower limit of -27.6, masculine values were below -27.6, and feminine values were above 27.6. Communication and expressiveness assessments were conducted by two speech-language pathologists (with a master’s degree and voice specialization) using the Speech Pathology Expressiveness Observation Guide (RoFOE), which contains 28 parameters divided into two subcategories: initial communication impact (6 parameters) and expressiveness (22 parameters), with the latter subdivided into vocal aspects (5), verbal aspects (10), and non-verbal aspects (7). The chi-square test was used to evaluate the frequency of communicative characteristics (p<0.05)..
Results: Voices of cisgender men and women were consistently perceived as masculine and feminine, respectively. Trans women’s voices were more frequently perceived as neutral, while trans men and non-binary individuals had varied distributions between masculine and neutral voice perceptions (p<.001). Regarding prosodic factors, voices perceived as feminine did not score in the question “Does any expressive characteristic dominate communication?” compared to neutral and masculine voices (p=0.046). Feminine voices were more frequently perceived as pleasant compared to neutral voices (p=0.012), which were also less frequently perceived with a natural pitch (p=0.028).
Conclusion: Cisgender individuals' voices were consistently perceived according to their self-identified gender, while transgender and non-binary individuals’ voices were more often perceived as neutral. Feminine-perceived voices were rated as more pleasant than neutral voices, and neutral voices were characterized by an unnatural pitch. Evaluating individuals with neutral voices remains challenging, even for experienced professionals.

Diego
Camilly
Ana Carolina
Martinho
Reis
Constantini