Friday Audiogram: Starkey Omega AI's DNN 360 Explained

What if your hearing tech could tell who you want to hear—even while you’re walking—and lift their voice above the crowd? We dive into Starkey Omega AI's DNN 360, a new approach that blends deep neural network noise management, intelligent directionality, and spatial awareness to make conversations stand out without turning the world into a dull hush. It’s built for real life, not lab silence, and it targets the moments that frustrate most: busy cafés, echoey lobbies, and outdoor chatter where voices blur together.

We break down how the system personalizes noise reduction using a model trained on speech, environmental sounds, and complex scenes, then pairs it with directionality that aims at the talker without collapsing the space around you. That balance yields measurable results—up to a 13 dB improvement in signal-to-noise ratio in diffuse noise and a reported 28% boost in speech understanding—while keeping listening natural and comfortable. The twist is motion: by integrating inertial measurement units, the device recognizes when your conversational partner is off to the side and adjusts focus as you move, solving the everyday mismatch between where you look and who you’re hearing.

We also explore spatial fidelity—minimizing interaural phase delays so your brain gets a stable, believable soundstage—because clarity isn’t just about suppression, it’s about preserving cues that help you separate sources. Together, these elements create a more effortless listening experience that adapts to your preferences instead of forcing you to adapt to it. If you care about hearing better in noise, you’ll come away with a clear picture of how AI, directionality, and motion sensing can work in concert to reduce effort and raise understanding.

Enjoyed the conversation? Follow the show, share this episode with someone who battles noisy rooms, and leave a quick review to tell us where hearing tech should go next.

Connect with the Hearing Matters Podcast Team

Email: hearingmatterspodcast@gmail.com

Instagram: @hearing_matters_podcast

Facebook: Hearing Matters Podcast

Chapters

• 0:00 Quick Kickoff
• 0:03 Introducing DNN 360 Hearing Tech
• 0:36 Real-World Gains In Noisy Places
• 0:45 Why This Isn’t Just Directionality
• 1:11 Sensors That Follow How You Move
• 1:36 Spatial Awareness And What Comes Next

Transcript

Dr. Dave Fabry: 0:00

This is the Friday audiogram. Let's go. This is the industry's first device that not only uses noise management that is optimized, personalized for the individual on the on the basis of the DNN training for speech and noise and other environmental sounds, but it also incorporates directionality. So this is the industry's first product that is incorporating the DNN with noise management that improves sound quality and comfort and ease of listening, plus directionality to get directly to that signal-to-noise ratio improvement, up to 13 dB improvement in ambient uh environments and diffuse noise. 28% on Michelle Hicks and her team measured 28% improvement in speech understanding, which is, as Brandon said, in noise, which is all it's what it's all about. The other element, though, someone could say, oh, well, we have adaptive directionality. Well, there's a very strong differentiator between DNN 360 with the DNN optimized for noise management, directionality plus uh the ability to use the IMUs, the inertial measurement units that we introduced into the hearing aid space in 2018, so that now uh in that same thing that would track the physical activity. Now we can start thinking, and Brandon, you said it, sometimes you're walking with someone on your right-hand side, and you may want to listen to them. Most of the time we're looking at what we want to hear, but there are occasions where now with the inertial measurement unit, as the person's walking along at 90 degrees relative to what's in front of me, I can see the DNN model training to adapt for that person that's moving alongside of me in tandem. So what DNN 360 does, spatial awareness through um uh matching and minimizing phase delays between the two ears, directionality and noise management, and now starting to think about the other integration of other sensors like the inertial measurement units. And that opens up the sky, is the limit, really, in terms of where we can go with personalized speech understanding, sound quality, and even movement integrated in and trained on the DNN model.