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

TL;DR

This review explores how recent large language and vision-language models can be leveraged to automate audio description generation, aiming to improve accessibility for visually impaired users while reducing human effort and costs.

Contribution

The paper provides a comprehensive review of current NLP and CV technologies applicable to automatic audio description generation and outlines key future research directions.

Findings

LLMs and VLMs show promise for automating AD generation

Current models can produce contextually relevant descriptions

Future research needed for improved accuracy and accessibility

Abstract

Audio descriptions (ADs) function as acoustic commentaries designed to assist blind persons and persons with visual impairments in accessing digital media content on television and in movies, among other settings. As an accessibility service typically provided by trained AD professionals, the generation of ADs demands significant human effort, making the process both time-consuming and costly. Recent advancements in natural language processing (NLP) and computer vision (CV), particularly in large language models (LLMs) and vision-language models (VLMs), have allowed for getting a step closer to automatic AD generation. This paper reviews the technologies pertinent to AD generation in the era of LLMs and VLMs: we discuss how state-of-the-art NLP and CV technologies can be applied to generate ADs and identify essential research directions for the future.