Could Audio-Described Films Benefit from Audio Introductions? An Audience Response Study

Abstract

Introduction Time constraints limit the quantity and type of information conveyed in audio description (AD) for films, in particular the cinematic aspects. Inspired by introductory notes for theatre AD, this study developed audio introductions (AIs) for Slumdog Millionaire and Man on Wire. Each AI comprised 10 minutes of continuous description incorporating information about the film's visual style, fuller descriptions of characters and settings, a brief synopsis, and cast and production details. The AIs were tested with participants who are blind and have low vision. Method Twenty-four visually impaired volunteers listened to the AIs before or after watching the films with AD, and gave feedback about their experience, either at sessions organized at the University of Roehampton, United Kingdom, or at home. Results This was a small-scale, exploratory study that showed a positive response to the concept of AIs for film. Most participants felt the AIs helped bring the films to life and made them easier to follow. The majority of participants wanted AIs for other films. Discussion AD guidelines discourage describers from mentioning camera work, yet most participants reported that this information in the AI was not too technical, and that the proportion of the AI devoted to visual style was about right (14 of 20 for Slumdog Millionaire, 14 of 16 for Man on Wire). Only a minority felt that the AIs “told me things I could find out for myself.” This suggests that access to screen media for people with a visual impairment can be enhanced by additional cinematic and other visual information. Implications for practitioners Given the limited time available for description during the film itself, AD providers should consider the use of AIs as a complement to standard film AD. These audio introductions could be stand-alone or accessed from a website.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 49 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.