![Matt Firth Matt Firth](https://tkt1957.com/wp-content/uploads/2024/09/52b71418-3156-44b4-8a23-cd9b8a32.png)
Audio Description plays an indispensable role in making media accessible to visually impaired individuals. But high production costs and significant time demands are a challenge that limits their availability and timeliness. By using AI, Globo has built a system to automatically narrate and audio mix a script created by an expert, reducing the need for a professional narrator and audio mixing team. This is particularly valuable on high turnaround productions, such as daily soaps. The system has been successfully running on real productions with impressive results. Our second paper, from the BBC presents details of a trial, held at Eurovision, implementing ADM (the open standard Audio Definition Model) for representing Next Generation Audio in an encoder agnostic way. For the live production, Serial-ADM and ADM Open-Source Control were successfully implemented and inter-operability established with a commercial encoder.
Speakers
![Pedro Leite](https://tkt1957.com/wp-content/uploads/2024/09/6d0de77b-61a6-4a82-94ce-9b4be14d.png)