Improving Dialogue Intelligibility in Streaming Media

Dialogue intelligibility is an increasing source of friction for consumers of streaming media [1][2][3]. Consumers expect dialogue to be intelligible for all content, in their listening environment, on all their devices, and for their individual hearing characteristics. Meeting these expectations is not a simple problem to solve [2]. When dialogue intelligibility becomes challenging the viewer has a poor user experience, leading to frustration and potentially content abandonment. Dialogue intelligibility can be impacted by multiple contributary factors across the entire delivery chain, from production, through distribution to playback. Dialogue is often masked by loud background music or sound effects, but adjusting the mix ratio can improve intelligibility [4][5]. This paper examines an innovative architectural approach to improving dialogue intelligibility that can be implemented at the streaming platform level, enabling improvements in dialogue clarity and offering different options to improve the overall audio experience for end users.