DTS Launches ‘Clear Dialogue’ Feature to Improve Dialogue Intelligibility

In early 2023, I watched an interesting YouTube video, made by Vox, called "Why We All Need Subtitles Now."  With the help of dialogue editor Austin Olivia Kendrick, the video explores the complicated reasons why dialogue in movies and TV is harder to make out than it used to be. Of course, your own experience depends upon a variety of factors, including the content you watch, the audio system you use, and your hearing ability. But according to Vox, most viewers use subtitles at least sometimes. When I watch with my mom and stepdad, both in their 70s, we use subtitles almost all the time. Now DTS is tackling this widespread issue with a feature called DTS Clear Dialogue, a new on-device solution that “leverages the latest advancements in AI-based audio processing to improve dialogue intelligibility for TV audio, delivering a more engaging and personalized audio experience for consumers at home,” according to the company.

DTS acknowledges that there are many reasons why viewers might have trouble understanding spoken onscreen dialogue, including limitations of the playback gear, inconsistent audio across content sources, environmental noise, and of course, differences in hearing ability. In fact, DTS turned to International Data Corporation (IDC), a global market intelligence/data provider, to learn directly from consumers about their needs and experiences watching content at home. IDC conducted a large consumer survey, led by senior research director Greg Ireland, who focuses on “the adoption of and engagement with consumer digital technologies, services and applications that transform consumer experiences, business models, and market opportunities,” according to IDC. He also has expertise in and provides in-depth analysis of the ways in which digital video content is distributed, consumed, and monetized. Xperi, the parent company to DTS, reported the results of the survey, revealing that 84% of the 1,200 U.S. adults who participated have experienced trouble understanding dialogue during TV shows and movies. More than 75% of survey respondents said they use captions/subtitles at least sometimes, with 30% reporting that subtitles are always or often turned on.

Sound quality is an essential part of the TV and video viewing experience. IDC’s consumer survey results continue to show that sound quality is highly important to a strong majority of viewers. Moreover, sound quality applies to older and young consumers, signaling that solutions that enable improved audio experiences address a highly important facet of the viewing experience for consumers across all generations.

— Greg Ireland, Senior Research Director at IDC

When viewers can’t hear the dialogue, it degrades the quality of their TV experience and turning on subtitles is a workaround, not a solution. To date, attempts to mitigate audio post-processing limitations don’t address the complex underlying problems across languages, content, and environment, and can’t account for each viewer’s individuality. Our device-side AI-based audio processing solution will allow TV OEMs to provide their viewers with the ability to customize and personalize their audio experience.

— Geir Skaaden, Chief Products & Services Officer at Xperi

As Xperi’s Geir Skaaden mentions in the quote above, DTS Clear Dialogue is unique in that it allows users to “personalize and optimize” their audio experience in order to improve their understanding of dialogue. To accomplish this, the feature reportedly uses AI audio processing innovation and machine learning techniques to “identify, separate, and enhance dialogue so listeners can better understand it.” DTS says that the feature works with all types of content, including movies, TV shows, sports, live events, news, and so on. Furthermore, Clear Dialogue is language-inclusive, meaning that it should work no matter what language is being spoken. DTS is offering Clear Dialogue as a feature for TV manufacturers to license and include in their products. Theoretically, it could also be applied in other AV hardware, from soundbars to mobile devices.

Some TV manufacturers, like Sony, have been working on their own in-house solutions. Sony’s “Voice Zoom 3” is a feature on select Sony BRAVIA TVs that uses AI machine learning to adjust the volume of human voices in relation to ambient audio. (This reportedly works especially well in conjunction with the Sony BRAVIA Theater Quad wireless home theater system, but the processing takes place inside the TV, so the Theater Quad can’t deliver this feature on its own — only when used alongside a compatible BRAVIA TV.) It remains to be seen how effective DTS’s technology is compared to its rivals, and whether TV manufacturers will embrace it — and be willing to pay to include DTS Clear Dialogue in their products. But what’s certain is that dialogue intelligibility is a growing concern among consumers, and one that manufacturers are all too aware of. Nearly 80% of the U.S. consumers who responded to Xperi’s survey said that DTS Clear Dialogue would increase their interest in a smart TV that offered it. About 40% said they’d use it because actors do not speak clearly (see the Pete Davidson example in the Vox video), while 38% complained that non-dialogue audio elements are too loud in the mix. Nearly 35% said that they watch in noisy environments, while just as many said that they have to keep the overall volume low in order to avoid disturbing others. In both of those cases, dialogue intelligibility suffers, but the underlying reasons are quite different.

According to DTS, understanding all of these varying situations has allowed the company to design a customizable system that can give users more control over their listening experiences. DTS Clear Dialogue identifies and virtually separates dialogue from other elements within an audio program, using AI to distinguish between dialogue, music, and special effects. It can then optimize the performance of these varying elements, according to the company. (At its core, this technology is similar to the AI-powered “demixing” techniques used to restore the audio in Peter Jackson’s Beatles documentary and in the “new” Beatles song Now And Then, released in 2023.) Users can “create personalized profiles with dialogue and non-dialogue settings, to address hearing impairment, listening environments, and audio preferences for varying conditions,” according to DTS. This level of customization is not something I’ve seen from competing solutions, and it could make a compelling argument for the widespread adoption of DTS Clear Dialogue.

This new level of AI-powered personalization addresses listener needs and can compensate for individual hearing impairment, device limitations, and listening environments for the first time. Providing this deeper level of understanding and enjoyment of TV content benefits everyone, from the viewers themselves to advertisers, streaming services, movie and TV studios, and TV OEMs.

— Geir Skaaden of Xperi

Conclusion

Like most Audioholics readers, my mom and stepdad don’t watch content using their Sony TV’s built-in speakers. They have an aging but perfectly respectable 5-channel system comprising a Denon receiver and speakers from NHT. What would really interest me would be a version of DTS Clear Dialogue that could work inside a typical AV receiver, or that the TV could apply to a multichannel digital signal before sending the audio out to an external system. DTS has not said whether any such version is in the works, but it certainly would not surprise me to see DTS Clear Dialogue adopted by soundbar manufacturers. Implementing the technology within a soundbar shouldn’t be too different from doing so inside a TV; in both cases, the engineers know the strengths, limitations, and performance parameters of the device’s built-in audio systems. Would the presence of DTS Clear Dialogue make you more interested in a particular TV, soundbar, or other device? Do you ever watch TV or movies with the captions turned on? Share your thoughts in the related forum thread below.