Over the past few years, discussions surrounding the comprehension of dialogue in film and television have grown louder. ABC has written several articles on the topic this year alone. There are some obvious films out there, such as anything from Christopher Nolan, as an example of pushing dialogue deep into the mix – most notably “Tenet”. While his intention for this is to feel more “realistic” and to help create uncomfortable moments, the backlash from the film may suggest he went a little too far. 

Now, with the introduction of Amazon Prime’s AI Dialogue Boost – are we heading toward a world where storytelling through the sound mix becomes irrelevant?  My argument is no, but that doesn’t mean the introduction of features like dialogue boost won’t be problematic. What’s Dialogue Boost?  Here’s a brief summary –  Amazon is introducing a new audio accessibility setting called Dialogue Boost, which uses AI to increase the volume of dialogue relative to the background music and effects, basically flattening the mix so that Dialogue is of a consistent level and the music and effect are lower. It’s intended to be an alternative to Closed-Capturing for people who are hard of hearing. 

While I agree that content should be consumed as intended, there should be considerations for accessibility features for individuals with audio impairments. Just as we have Audio Description or contrast controls for people with visual impairments, it only makes sense to provide functions for those with auditory impairments. The question is: Should we rely on AI or explore other models?

As someone with APD, I personally hate subtitles, as it draws away from the visuals and means that I have more information I have to process, which takes a lot of brain energy for someone like myself with a processing disorder. If I can’t watch content in a sound control room (ie my sound edit suite), I normally will live with the fact that I might not understand everything that is being said. When I consume media at home, I’ll tend to lean toward games, as the sound mix can be adjusted, with different levels for Dialogue, Effects, and Music. More recent games have even more controls for key or story dialogue & effects vs background. The flexibility to be able to reduce the music and effects (even by a notch or two), makes a world of difference to how I consume games. So can this idea of adjustable stems be an alternative to Dialogue Boost?

Ultimately, we need to find a balance between preserving the filmmaker’s artistic vision and ensuring accessibility for all. While AI dialogue boost is addressing an issue, the ability to customize audio settings using stems would provide a more tailored and inclusive experience while preserving a lot of the intended Mix. I’m not saying this is the answer, but to me, this seems like a better solution – and with elements that we already have to create with most deliverable specs. Plus anything that isn’t altering creative works using AI is a bonus!

Articles Mentioned:
ABC: Here’s why film and TV dialogue has got harder to hear
ABC: Subtitles become popular among general population
Prime Video launches a new accessibility feature