Immersion in audio description. The impact of style and vocal delivery on users’ experience
Year of publication
Audio description (AD) is no longer seen as an exclusive accessibility service dedicated to a small fraction of individuals, but as an inclusive service, addressing not only the needs of people with sensory impairments, but also those of other audiences, including the elderly or children. The demand for this service is likely to grow in the near future. Although a drive for quantity of accessible audiovisual products can be observed on the market, the issue of the quality of user experience should not be overlooked. Framed within the inclusive model of accessibility, this PhD thesis aims to explore the impact of AD on the emotional response of its target audiences – blind and visually impaired (B/VIP) users. To this end, two research studies were carried out: one focusing on AD style and the other on AD vocal delivery. The rationale behind selecting these parameters was related to the findings of previous research conducted in the field. Two main research objectives were set. The first one, concerning AD style, centred around studying the reception of two AD styles – standard and creative – by B/VIP users. The second one, linked to AD vocal delivery, involved studying the reception of two AD voice types – human and synthetic – by B/VIP users for two genres: fiction and documentary. The novel aspect of this PhD thesis is the methodological approach adopted. It follows a user-centric methodology, not limiting itself to reporting on users' opinions, but also aiming to measure their emotional response to a stimuli presented. The research tool that is used to gauge user experience is the Independent Television Commission Sense of Presence Inventory questionnaire, one of five canonical questionnaires used for measuring presence. Two additional items – interest and confusion – are added for users to report their emotional response to the stimuli shown. The results indicate that not only AD scripts – creative or standard – but also AD delivery voices have a direct effect on the reception of a film, and therefore on users' experience. When it comes to AD style, the results show that creative AD, compared to standard AD, yielded higher levels of presence for all participants. Overall, the creative AD style seemed more natural, especially to participants with recent sight loss. However, it turned out to be more confusing than standard AD, possibly because the participants were not used to cinematic terminology included in the script or did not expect to hear unsavoury vocabulary. Nevertheless, when explicitly asked which AD style – creative or standard – they prefer, the majority of participants opted for creative AD. When it comes to AD vocal delivery, AD narrated by a human, compared to text-to-speech (TTS) AD, prompted significantly higher levels of presence for fiction. Presence rates for documentary were similar, with no statistically significant differences in relation to AD voice type. Fiction with a human voice AD was assessed by participants as being more interesting and less confusing than fiction with TTS AD. In the case of documentary, the levels of interest and confusion were comparable for both voice types. When asked which voice type they prefer for a given genre, participants showed a preference for AD read by a human for fiction, but no preference between a human voice AD and TTS AD for documentary. Overall, the findings suggest that when properly delivered, both in terms of style and voice, AD may increase the chances of B/VIP audiences having a more engaging viewing experience. It is hoped that this PhD thesis will act as a useful stepping stone towards understanding the links between the AD product and user context, and towards conducting further research in this field.