Cues for Perception of Gender in Synthetic Voices and the Role of Identity

Maxwell Hope, Jason Lilley


Perception of gender in voice is not an under-researched area. Previous studies have been conducted in the hopes of pinpointing what aspects of voice (e.g. fundamental frequency, intonation, etc.) carry the largest cues for skewing gender perception. These studies have to date been conducted within the framework of the gender binary, i.e. men’s vs. women’s voices, which have left out the exploration of perception of something besides simply femininity and masculinity.

The literature thus far has not endeavored to keep pitch in the “androgynous” zone while manipulating other aspects such as the F0 contour or other acoustic parameters. Additionally, past literature on speech perception has neglected to explicitly include members of the gender expansive community. Hence, we recruited participants of all genders and first sought to identify cues for gender perception in synthetically made voices and then examine the relationship between one’s own sense of gender identity and the perception of gender in synthetically made voices for native speakers of American English. We found that vocal tract acoustics are most important for swaying perception of gender and one’s own gender identity influences gender perception in voice.


 DOI: 10.21437/Interspeech.2020-2657

Cite as: Hope, M., Lilley, J. (2020) Cues for Perception of Gender in Synthetic Voices and the Role of Identity. Proc. Interspeech 2020, 4143-4147, DOI: 10.21437/Interspeech.2020-2657.


@inproceedings{Hope2020,
  author={Maxwell Hope and Jason Lilley},
  title={{Cues for Perception of Gender in Synthetic Voices and the Role of Identity}},
  year=2020,
  booktitle={Proc. Interspeech 2020},
  pages={4143--4147},
  doi={10.21437/Interspeech.2020-2657},
  url={http://dx.doi.org/10.21437/Interspeech.2020-2657}
}