Shaping the visually impaired perception of gender through Netflix audio description of movies and series

Abstract This paper examines how different gender identities are treated in the audio description feature provided on the Netflix platform. It examines how gender is reflected in language through the use of gendered terms and pronouns and how these are determined and assigned. The paper analyses if gender is assigned based on the visual information provided or if extra resources are used to confirm the validity of the descriptions. A set of Netflix movies and series with diverse casts were thus examined to compare how gender is treated in descriptions. The analysis searches for the existence of bias and assumptions and how they affect the description and described groups. The findings revealed that basing descriptions on visuals creates confusion between gender expression, performance, and gender identity, which could lead to the misgendering of both cisgender and transgender characters. Thus, it is best not to rely on stereotypes and assumptions and to research the characters through textual and extratextual information to properly describe them. Examining the uses of gender-neutral and gender-fair language in contrast with gendered labels revealed that it can be used to avoid misgendering but is currently mainly used as a tool to enhance mystery.