Abstract:Audio description (AD) is one of the main methods that people who are blind or low vision (B/LV) use to access film, television, and theatre content. AD is a second audio track inserted into the space(s) where speech is absent, which tends to be only a few seconds. Contained in that second track is an audio description of the important visual information contained within a specific scene. However, as there is insufficient time to describe all visual information, decisions about what is important to describe and how to present that information (style) to optimize a B/LV viewer's entertainment experience are required. Most research to date has considered only short-term, single-episode experiences to gauge viewers' reactions to the AD content. In addition, this research typically has used a monotone, single style of audio description, which is defined as "the conventional style" in this paper. We use an integrative style instead, that is defined as 'AD designed to fit a specific show", and differed between shows. We carried out a within-subjects longitudinal study with eight episodes of a dark comedy, using different description styles and describers in order to assess viewer engagement and preferences for AD describer style, language use, timing, and fit to the show. Twenty-four blind participants viewed and rated all eight episodes. Major findings included that most participants found the integrative style entertaining, a fit with the specific episodes, and enjoyable. Some participants, however, preferred the conventional style and struggled with the language and topic of a dark comedy and its associated descriptions.

Describe Now: User-Driven Audio Description for Blind and Low Vision Individuals

Audio Description Customization

Toward Automatic Audio Description Generation for Accessible Videos

Audio Description Generation in the Era of LLMs and VLMs: A Review of Transferable Generative AI Technologies

Making Accessible Movies Easily: an Intelligent Tool for Authoring and Integrating Audio Descriptions to Movies

NarrationBot and InfoBot: A Hybrid System for Automated Video Description

"It's Kind of Context Dependent": Understanding Blind and Low Vision People's Video Accessibility Preferences Across Viewing Scenarios

SPICA: Interactive Video Content Exploration through Augmented Audio Descriptions for Blind or Low-Vision Viewers

Investigating Use Cases of AI-Powered Scene Description Applications for Blind and Low Vision People

Impacts on quality: Enjoyment factors in blind and low vision audience entertainment ratings: A qualitative study

NowYouSee Me: Context-Aware Automatic Audio Description

AutoAD: Movie Description in Context

You Described, We Archived: A Rich Audio Description Dataset

AutoAD-Zero: A Training-Free Framework for Zero-Shot Audio Description

AutoAD III: The Prequel -- Back to the Pixels

LLM-AD: Large Language Model based Audio Description System

AutoAD II: The Sequel -- Who, When, and What in Movie Audio Description

Contextual AD Narration with Interleaved Multimodal Sequence

Video Description Generation Using Audio and Visual Cues

Audio describing the mental dimension of narrative characters. Insights from a Flemish case study.

WorldScribe: Towards Context-Aware Live Visual Descriptions