The Use of Artificial Intelligence Enabling Scalable Audio Description on Brazilian Television: A Workflow Proposal

Abstract

Recently, Artificial Intelligence (AI) technologies have been gaining ground in various areas of knowledge, significantly impacting many academic and business spheres. One application that can benefit from AI is the inclusion of people with disabilities in audiovisual content, where the scaling capacity of certain processes can bring new accessibility opportunities. In this work, we show what a traditional workflow of an audio description for dramaturgy audiovisual content looks like, and from there, we propose a new workflow for generating audio description audios for visually impaired people using synthetic voice created with Artificial Intelligence models. The proposed workflow simplifies and considerably reduces production time and costs, besides allowing the generation of audios on a larger scale compared to a traditional workflow, enabling a broader reach of the target audience. It also allows multiple people to work simultaneously on the same project while preserving sound identity through the synthetic voice and standardized mixing. With this proposal, we believe that accessibility on Brazilian television can be expanded to serve a much larger audience.