AI hearing aids separate ‘haves’ from ‘have-nots’

Premium-priced hearing aids just keep getting better and better. That’s great news for those who can afford them. But what about everyone else?

This year’s introduction of hearing aids with true artificial intelligence (AI) capabilities have signaled a great leap forward in the quest for the hearing industry’s holy grail—technology that enables people with hearing loss to understand speech in noise as well as people with normal hearing. But as with past advances in hearing technology, the new high-performance hearing aids once again widen the gap between the “haves” and the “have-nots” in the hearing loss population.

The “haves” are the relatively few consumers with the means to shell out thousands of dollars from their own pockets for hearing aids with all the bells and whistles. The “have-nots” are the tens of millions of people with hearing loss worldwide who can’t even afford to spend a few hundred dollars on better hearing, let alone thousands.

 OTC hearing aids are slow out of the gate…

Affordable, low-cost over the counter (OTC) hearing aids, sold without requiring a prescription or a visit to a hearing care professional, are intended to improve the lot of the have-nots. But they have not taken consumer markets by storm, at least not yet.

Apple recently got huge consumer media attention when the US Food and Drug Administration approved the AirPods Pro 2 as FDA-approved hearing aids. But the AirPods are bulkier than traditional in-the-ear hearing aids, and their batteries only last about five hours, versus standard hearing aid batteries which enable all-day wear. And beyond Apple, adoption of other affordable hearing aid options has been slower than Congress hoped when it passed legislation—way back in 2017—enabling over-the-counter sales of non-prescription hearing aids.

 Meanwhile, a series of less-heralded announcements from the world’s largest premium hearing aid brands have made a more significant immediate impact on hearing aid performance—and on global hearing aid markets—than the emerging class of OTC hearing aids.

…while AI turbocharges high-end hearing aids

As in many other realms, artificial intelligence technologies are poised to transform the hearing aid industry. In 2024, top hearing aid brands have announced new AI sound processing platforms powered by deep neural networking (DNN) technology. The new hearing aids are delivering remarkable improvements in comprehension of speech in noise.

Deep neural networks for hearing aids are trained on huge datasets containing tens of thousands of various noise types ranging from crowded areas outside to indoor restaurant or cocktail party noise to environmental noise such as wind or loud traffic. DNN-based models learn complex patterns in speech and noise, making it possible to identify speech tones in many environments to separate speech from background noise. And DNNs can filter out noise while recognizing and preserving speech nuances, such as emotional tone or accents, which were often lost in older technologies.

AI sound processing platforms driven by DNN technology isolate speech and improve the signal-to-noise (SNR) ratio in the sound delivered to the hearing aid wearer. As such, they can dramatically enhance the ability to understand conversations even in challenging listening environments.

But time, effort, and massive computing power are required to train DNN models for hearing aids, amounting to huge investments for hearing aid developers. And new semiconductor chips with enough processing power to incorporate AI into hearing aids are extremely expensive to develop and manufacture. Those investment costs translate to consumer prices for a pair of AI hearing aids that can start at and most often rise far north of approximately $3,500.

Top hearing aid brands lead the way in AI

AI has become a ubiquitous marketing buzzword that’s often used by consumer brands selling barely intelligent products. But in the hearing aid industry, true AI is here. In 2024, three of the world’s leading hearing-aid manufacturers have launched sound processing platforms with honest-to-goodness AI capabilities that are delivering next-generation improvements in comprehension of speech in noise. 

• Starkey Edge AI from Starkey Hearing Technologies fully integrates advanced neural processing circuits onto the hearing aid’s sound processing chip, delivering higher processing speeds, longer battery life, and a smaller footprint. Using Starkey’s DNN model, the new platform delivers a 30% improvement in identifying speech compared to Starkey’s previous hearing aids. And its rechargeable batteries last 51 hours. The hearing aids also support the latest Bluetooth LE Audio technology, including Auracast™ broadcast streaming, allowing seamless connectivity to a wide range of devices.
• Phonak Audéo Sphere Infinio hearing aids from Sonova Group’s Phonak brand incorporate real-time AI powered by two new sound processing chips. A DNN trained on over 22 million sound samples, enables precise, dynamic adjustments to help users understand speech from any direction. Phonak says the Infinio platform offers a 10-decibel improvement in speech-to-noise ratio (SNR) and reduces listening effort while providing connectivity with Bluetooth-enabled devices.
• Oticon Intent hearing aids from Demant A/S build on Oticon’s “BrainHearing” technology, first introduced in 2014, designed to reduce cognitive load by providing access to a balanced soundscape. Now a decade later, Oticon has added more AI to the mix with its second generation of DNN model, trained on a wider variety of complex sound environments. The Oticon Intent sound processing platform also features multiple “user-intent” sensors that monitors head movement, body activity, conversation, and the acoustic environment to adjust sound processing based on the user’s communication goals.

These advances in the core benefit delivered by hearing aids—the ability to better comprehend speech in noisy environments—will widen the performance gap between premium hearing aids for the haves and entry level OTC hearing aids for the have-nots.

But that doesn’t mean the have-nots will be permanently left out in the cold. In fact, given Moore’s law, which holds that the number of circuits on a semiconductor chip doubles every two years at the same or lower cost, it won’t be long before true AI processing capabilities are within the reach of lower cost OTC hearing aids.

AI breathes new life into old premium hearing aid brands

The manufacturers of premium prescription hearing aids for years have been hearing the footsteps of lower cost OTC competitors that are rapidly improving in performance. Lower cost hearing aids offering performance that nearly equals the premium brands have put price and margin pressure on the leading manufacturers. Look back over the annual financial reports from the major premium hearing aid manufacturers and you will see that “ongoing reductions in “ASP”—lower average selling prices, which translate to lower profits—are a perennial concern.

But every time lower-priced competitors seem to be closing the gap. the premium hearing aid manufacturers have always proven adept at staying one step ahead with new capabilities and better performance to justify their higher prices. When simple amplifiers started to replicate professional-grade hearing aid amplification, the hearing aid manufacturers introduced innovations such as directional microphones that localize and enhance a speaker’s voice. When low-end manufacturers introduced hearing aids with inexpensive directional microphones, the premium brands started offering Bluetooth streaming of phone calls and audio.

Now, at a time when OTC hearing aids are starting to offer Bluetooth connectivity and streaming, AI has breathed new life into the premium brands. They are staying a step ahead again with next-generation improvements in comprehension of speech in noise enabled by AI and DNN processing. 

How will OTC hearing aid providers close the gap? 

Now that the premium hearing aid manufacturers have raised the bar once again, will suppliers of entry-level OTC hearing aids be able to close the gap? Most likely, yes.

Among other things, AI may well improve the performance of the self-fitting software required for the success of OTC hearing aids. Self-fit apps for OTC hearing aids have come a long way, but they still have a long way to go before they deliver amplification mapped to your personal hearing profile as well as an audiologist fitting. But AI may enable the industry to reach that goal faster. Here is what Andreas Thelander Bertelsen, who pioneered development of AI hearing aids for Oticon, ReSound, Whisper, and Meta, told HearingTracker’s Karl Strom:

"AI will enable us to build new kinds of interactive experiences that allow both providers and patients to reach a more personalized outcome in less time using the data collected. In the OTC space, I think the trend will be toward simpler yet more interactive software for consumers to do self-fittings. The questionnaires and pure-tone tests that hearing care professionals rely on today will evolve into interactive games or augmented reality (AR) experiences that will be able to transport the users to a complex sound environment right in their homes. Instead of guessing how something might sound, you’ll be able to try it immediately by listening. These experiences and preferences, in turn, can then all be collected and analyzed by AI to provide much more precise personalization."

             -- Andreas Thelander Bertelsen, Audio Lead for Health Tech, Meta

And there’s no doubt that the more affordable OTC hearing aids will continue delivering improvements in sound processing performance. As the semiconductor industry increases the density and reduces the cost of sound processing chips, affordable OTC hearing aids will see rapid improvements in performance. And as the cost of training DNN models with complex soundscapes comes down, OTC hearing aid manufacturers will be able to start delivering true AI capabilities.

Constant improvement

It’s been seven long years since Congress passed the OTC hearing aid act. And we’re still in a market where a few “haves” are enjoying better hearing enhancement than ever, while the “have-nots” continue to struggle with entry-level hearing solutions that never quite hit the bullseye. 

But we’ve seen constant improvement at the same time. Premium hearing aid manufacturers continue delivering better hearing enhancements than ever. And the OTC makers continue improving the performance of their product and self-fitting solutions at lower costs than prescription hearing aids.

And while AI and DNN processing currently have widened the gap between the “haves” and “have-nots” once again, don’t be surprised if those technologies quickly come together for OTC hearing aids in a way that will finally enable them to start closing the gap.