Sonos Introduces AI Speech Enhancement for Hearing Accessibility

If you’ve found yourself cranking up the volume or switching on subtitles to catch what characters are saying on screen, you’re not alone. Muffled dialogue has become one of the most common frustrations in modern film and television viewing, and it’s a particularly acute problem for those with hearing loss.

Now, Sonos is tackling the issue head-on with the debut of its new AI-powered Speech Enhancement tool. Rolling out this week as a free software update for the Arc Ultra soundbar, the feature lets users fine-tune dialogue clarity across four selectable levels, including one specifically designed for people with hearing difficulties. It’s part of what the company says is a broader commitment to inclusive design and accessible home audio.

Developed in collaboration with the Royal National Institute for Deaf People (RNID) and award-winning film sound mixer Chris Jenkins, Speech Enhancement goes beyond traditional “dialogue boost” settings. Using AI and machine learning, the tool actively separates spoken words from other elements in a film or show’s centre channel audio, clarifying them in real time without overcompensating overall volume or distorting the original mix.

The result is a system that doesn’t just make dialogue louder — it makes it clearer. Sonos Principal Research Scientist Matt Benatan said the feature is more than technical precision:

It’s not just practical, it's emotional… If one viewer cannot fully hear the dialogue, their ability to enjoy and participate in the moment can be lost.

Sonos Sound Experience Engineer Harry Jones added that machine learning allowed the team to "draw out just the dialogue at the most needed times" without compromising the holistic cinematic experience. The team’s close work with RNID ensured the feature was grounded in real-world feedback, gathered from nearly a year of testing with 37 participants of varying ages and hearing capabilities.

Users can access the new feature directly from the Sonos app, where they’ll find four selectable levels of enhancement:

• Low: Subtle tuning that gently lifts dialogue without interfering with the original sound mix.
• Medium: A balanced setting improves speech clarity while preserving mix integrity.
• High: More assertive enhancement that prioritises dialogue, reducing background audio distraction.
• Max: Designed specifically for people with hearing loss, this setting makes dialogue dominant and compresses surrounding audio dynamics to keep voices intelligible even in busy scenes.

Chris Jenkins, whose credits include blockbusters like Mad Max: Fury Road, worked closely with the Sonos team to fine-tune the system, ensuring that while dialogue takes priority, the emotional weight of music and action sequences remains intact. He commented:

It’s a huge step forward, it’s also a testament to the importance of retaining a human touch when building with AI.

From the beginning, the Speech Enhancement project was shaped by the lived experiences of people with hearing challenges.

RNID researcher Lauren Ward highlighted the broader implications:

One in three adults in the UK experience hearing loss, and just under one in four in the US. This tool has the potential to impact a large number of people, including those who may not even realise their hearing is changing.

According to Sonos, the result is a feature that not only enhances accessibility but redefines what a home theatre experience can be—thoughtful, inclusive, and responsive to real human needs.

Speech Enhancement is now available as part of a free update for all Arc Ultra users.

Visit Sonos for more information