Meta's smart glasses are about to do something that sounds a little strange until you look at how they actually work: they can be used as a hearing aid.
Meta calls the feature hearing enhancement, and says it is coming to supported AI glasses in the United States later this year. It will cost $149.99 to add, or come with a Meta One subscription. (Meta Newsroom)
But there is a difference between saying glasses can make sound louder and saying they are a hearing aid.
In this case, Meta is not simply adding another audio setting. The FDA cleared the underlying software as an over-the-counter, self-fitting hearing aid for adults with perceived mild to moderate hearing loss.
So yes — Meta's glasses really can function as a hearing aid.
The catch is that the clearance is much narrower than the headline makes it sound, and almost all of the detail is in a document most coverage never opened.
What did the FDA actually clear?
The FDA record is unusually useful here, because it says things Meta's announcement does not.
The device submitted by Meta is called Hearing assist. It is classified as Air-Conduction Hearing Aid Software under regulation 21 CFR 874.3335, product code SCR, regulatory class II. The applicant is Meta Platform Technologies, LLC, of Burlingame, California.
It was cleared under 510(k) number K254044, with a decision date of July 13, 2026. The decision is listed as "Substantially Equivalent" — a 510(k) clearance, not an FDA "approval" in the colloquial sense. (FDA 510(k) database)
That distinction is not pedantry. It means the FDA determined Meta's device was sufficiently similar in intended use and technology to a device already legally on the market. And the FDA names that device in the filing.
The predicate is Apple's Hearing Aid Feature — the AirPods Pro 2 feature cleared under DEN230081, a De Novo classification. Meta's filing compares the two directly, and concludes the change of hardware platform from earbuds to glasses "does not raise any new safety or effectiveness questions." (K254044 510(k) summary)
The filing also notes a precedent: a glasses form factor with microphones and an external speaker had already been cleared by the FDA under the same SCR product code, in the Nuance device K243150.
Here is the whole thing in one table.
| Claim | What the evidence says |
|---|---|
| Meta glasses can act as a hearing aid | True, for the cleared use |
| They are FDA cleared | True |
| They work for everyone with hearing loss | False |
| They are intended for mild to moderate loss | True |
| An audiologist has to fit them | No — they are self-fitting |
| You need a prescription | No — over the counter |
| They currently work with Android | Not in the cleared configuration |
| They replace prescription hearing aids | No |
| Meta tested the self-fitting system | Yes |
| The FDA approved the glasses themselves as a universal hearing device | No |
The rest of this article is the reasoning behind those rows.
How do Meta's glasses actually become a hearing aid?
The system is two pieces of software working together.
One runs on the compatible smart glasses. It is firmware, installed and running on the glasses platform.
The other runs on a smartphone — and in the cleared configuration, that means a compatible iOS phone. The filing describes the device as a pair of software modules: the Hearing assist Smartphone iOS Application, and the Hearing assist Smart Glasses App.
Before the amplification does anything, the wearer goes through onboarding and self-fitting in the phone app while wearing the glasses. That process builds a hearing profile.
The profile is a proprietary data format that specifies how real-world sound should be amplified for each ear. Once it exists, the wearer can switch on real-world sound amplification.
The glasses then use the platform's own hardware — a multiple-microphone system and the built-in speakers — to pick up the sound around the wearer and deliver amplified audio back to the ears.
Nothing sits inside the ear canal.
That is the biggest physical difference between this and something like AirPods Pro. With AirPods, the amplified signal leaves an ear tip positioned inside the ear canal. With Meta's system, the signal arrives from speakers in the glasses, outside the ear entirely.
The filing points out that the glasses platform is a combination product: already a class 1 device for vision correction, plus a general computing platform with cameras and microphones. The hearing software borrows that general-purpose hardware rather than adding a dedicated hearing device.
The hearing test isn't just a regular hearing check
The setup process is part of the medical device itself, not a separate screening app.
The filing is explicit that the hearing test and profile generator are built into the Hearing assist software modules and are not a standalone application. Users have to complete onboarding and self-fitting before they can use the device at all.
The app then gives the wearer three kinds of adjustment: volume, sharpness and balance.
There is an interesting limit buried here. Volume and sharpness are applied bilaterally — to both ears at the same time. The user cannot fine-tune those two independently for each ear.
The device still protects each ear separately: the effective gain in each ear is bounded by per-ear gain limits, described in the filing as ensuring safe amplification in both the better and the poorer ear.
So the software is doing more than switching on a generic "boost voices" mode. It is building a per-ear amplification map and then constraining it, which is a different job from a simple loudness setting.
How well did Meta's system work?
This is where the filing gets much more interesting than the announcement.
Meta tested Hearing assist in a clinical investigation involving 142 subjects, aged 18 and over, with mild to moderate hearing impairment or perceived hearing difficulties despite normal hearing thresholds. Participants were enrolled by four-frequency pure tone average across three bands: no impairment with perceived hearing loss (15–25 dB HL), mild (26–40 dB HL) and moderate (41–60 dB HL).
The study ran at three clinical sites across the United States, in two phases. Phase one was lab-based, where the devices were fitted and the clinical assessments were done. Phase two sent participants into their own homes and communities for 14 days, using the device in natural listening environments.
It compared two groups. One self-fitted the device. The other had it set up and fitted by a licensed hearing care professional.
The point was never to prove Meta's software beats a professional. It was to find out whether self-fitting gets close enough, and the endpoints were built around non-inferiority.
The primary endpoint was the International Outcome Inventory for Hearing Aids (IOI-HA) after the 14 days of field use. The self-fit group scored 25.77 on average; the professional-fit group scored 26.81. The mean difference was −1.04, with a 95% confidence interval of −2.28 to 0.24 and a p-value of 0.004 for the non-inferiority test. The null hypothesis was rejected: self-fitting was found non-inferior.
The secondary endpoint measured objective benefit with the Quick Speech-in-Noise test. Mean SNR loss was 4.4 dB for the self-fit group and 4.2 dB for the professional-fit group, a difference of −0.2 dB with a 95% confidence interval of −0.5 to 0.3 and a p-value below 0.001. Non-inferior again.
There was also a within-participant measurement, and this is the number worth remembering: participants showed an average signal-to-noise benefit of 4.8 dB when using Hearing assist, compared with their own performance without it. That is the honest version of "it helps you hear in noisy places." It is measured against not having the device, not against a rival product.
Real-ear measurements confirmed equivalent insertion gain between the two groups at 50, 65 and 75 dB SPL, consistent with the survey and speech-in-noise results.
What this does not tell you is whether you personally will follow conversation across a loud restaurant. The study gives us measurable, well-designed evidence of benefit in the population and conditions tested. That is a real result, and it is a narrower claim than "hearing solved."
What about safety?
The clinical study reported no unanticipated adverse device effects and no serious adverse device effects.
There were 6 adverse device effects reported by 5 of the 142 participants. Headache was the most common, with 5 reports. There was 1 report of tiredness and fatigue.
The filing describes those as consistent with the expected safety profile of a non-significant-risk device: mild and transient, with all subjects recovering without any intervention, and nobody withdrawing from the study because of them. Headache and fatigue are, as the filing notes, anticipated when someone first starts using a hearing aid and acclimates to it.
There were also 4 serious adverse events in 2 of the 142 participants, which the filing states were unrelated to the study device or procedures — one fall with injury and one cardiac event, both requiring hospitalization. One participant withdrew before completion, and five protocol deviations were reported.
So the evidence does not show a safety problem with the hearing software itself.
It also does not mean that anyone with trouble hearing should diagnose themselves and buy a pair of glasses. That is a different question, and it is the one the FDA's own hearing-aid guidance answers.
The biggest limitation: mild to moderate hearing loss
This is the part that tends to disappear when the story gets compressed into "Meta glasses are now hearing aids."
They are not intended for every level of hearing loss. The FDA's over-the-counter hearing aid category, which began on October 17, 2022, exists for adults 18 and older with perceived mild to moderate hearing loss.
The agency is direct that these devices are not for severe or profound hearing loss, because OTC hearing aids are limited in their maximum output and may not give adequate benefit for more severe losses. Meta's clearance sits inside the same boundary.
There is a second limitation that has nothing to do with volume. Sometimes the problem is not that sound needs amplifying.
The FDA lists specific situations where a doctor — preferably an ear, nose and throat specialist — should be involved instead: an unusual ear shape or an ear deformed in an accident; blood, pus or fluid coming from the ear in the past six months; an ear that is painful or uncomfortable; a lot of ear wax or the feeling that something is in the ear; real dizziness or vertigo; hearing that changed suddenly in the last six months; hearing that gets worse and then better again; worse hearing in one ear; or ringing and buzzing in only one ear. (FDA: OTC Hearing Aids)
A pair of smart glasses cannot tell you why your hearing changed.
It can only amplify sound within the limits of the device.
Are these really better than traditional hearing aids?
Not necessarily. They are different.
A professionally fitted hearing aid is selected and programmed around a specific person's hearing loss, and a professional can also notice problems a self-fitting consumer device will never identify.
Meta's system is designed to make hearing assistance easier to obtain. The announcement sells exactly that: set up at home in minutes, no clinic and no prescription. (Meta Newsroom)
That has real value for someone with mild difficulty who does not want an appointment, does not want to wear a traditional hearing aid, and does not want anything inside the ear.
And the glasses do not look like hearing aids. They are glasses — which matters for someone who already wears glasses every day.
But convenience is not the same thing as universal suitability, and the study behind this clearance tested a specific population with a specific degree of loss.
If you are weighing glasses against prescription hearing care, it is also worth reading why earbuds stay in or fall out — the ear-canal for Meta's approach is the point: the glasses do not occupy it at all.
How is this different from AirPods Pro?
The comparison with Apple is not a coincidence. Meta's FDA filing names Apple's Hearing Aid Feature as its predicate device, and compares the two at length.
Both are over-the-counter, self-fitting hearing aid technologies for adults with perceived mild to moderate hearing loss. Both use wide dynamic range compression. Both include active noise reduction and remote firmware updates through a phone app. Both are cleared under the same product code and the same regulation.
The difference is where the hardware sits.
AirPods Pro deliver amplified sound through an ear tip inside the concha and ear canal. Meta delivers it from speakers in the glasses, without obstructing the ear canal.
On the bench, the two are close: Meta's device reaches a maximum output of 108.6 dB SPL where Apple's reaches 105.93 dB, and Meta measured a maximum self-generated noise of 25.2 dBA against Apple's 28.2 dBA. The frequency responses differ — 232 Hz to 7,809 Hz for the glasses against a wider 100 Hz to 10,000 Hz for the earbuds.
The filing's position is that the hardware difference was evaluated with bench and clinical testing and "does not raise any new safety or effectiveness questions."
So the interesting comparison is not simply which one is better. It is whether you would rather have hearing assistance in something you wear on your face or something you wear in your ears.
What about Android?
For now, the FDA filing points to an iOS smartphone as part of the cleared system.
This is easy to miss, and it is worth stating plainly: the glasses are not an independent hearing aid that works without a phone. The cleared configuration is the smart glasses firmware plus a companion application running on a compatible iOS phone. Onboarding, self-fitting and the profile all happen there.
Android is not simply missing. It is pre-cleared as a planned change.
The FDA's clearance letter states that the substantial equivalence determination included review and clearance of a Predetermined Change Control Plan under section 515C of the Food, Drug and Cosmetic Act. Under that mechanism, a change consistent with an established plan does not need a new 510(k).
Meta's plan is specifically for adding an Android mobile platform to the Hearing assist smartphone app, with defined testing: no change to input type, output type, other modules or intended use, verified against product requirements plus cybersecurity testing for the Android platform.
That is a stronger position than a vague promise, and it is still not the same thing as Android being available today. The cleared and available product is the iOS configuration. If you are on Android, the sensible answer is to wait for Meta to announce actual availability rather than assuming the feature will work on day one.
Which Meta glasses will support it?
This is another place where the announcement leaves an open question.
Meta describes hearing enhancement as coming to supported AI glasses. It has not published a complete list of every compatible model in the announcement itself.
That matters if you are thinking about buying glasses specifically for hearing assistance.
Speakers and microphones are not enough on their own. The filing says the software was tested with "representative, compatible computing platform hardware," which tells us the platform has to meet the hearing system's requirements — not which pairs of glasses will qualify at launch.
Meta has been expanding the lineup: the announcement covers new Meta Glasses colourways, a Kylie Jenner edition, the first LISA collaboration, and an entry-level pair, alongside the separate Meta VR Glasses that Meta says weigh about 100 grams. That is a widening hardware family, which makes a compatibility list more necessary, not less.
So do not assume compatibility from the product name. Wait for the supported-model list.
What does the $149.99 actually buy?
Meta says hearing enhancement will be available to add to supported AI glasses for $149.99, or come with a Meta One subscription, launching in the US later this year. Meta also says it is eligible for HSA and FSA reimbursement, subject to your plan. (Meta Newsroom)
The important part is what the number is not. It is not the price of the glasses.
You still need compatible glasses. So the real cost of getting Meta's hearing assistance is:
compatible Meta glasses + the hearing enhancement feature.
That changes the comparison considerably. It is also why lining $149.99 up directly against the price of a complete hearing aid is misleading — the glasses are doing several other jobs at once, and you are not buying them purely as a medical device.
So, can Meta glasses replace hearing aids?
For some people, they may remove the need for a separate over-the-counter hearing-aid device.
For everyone? No.
The evidence supports a much narrower answer than the headline. Meta's hearing enhancement is a genuine FDA-cleared OTC hearing aid. It is intended for adults with perceived mild to moderate hearing loss, it builds a self-fitting per-ear hearing profile, and its clinical study found self-fitting non-inferior to professional fitting inside the studied population — with a measurable average 4.8 dB signal-to-noise benefit versus not using the device.
That is a legitimate hearing-aid function.
But the clearance does not turn a pair of glasses into a universal substitute for prescription hearing aids or for professional hearing care. Someone with severe loss needs something different. Someone whose hearing changed suddenly, or who has pain, drainage or dizziness, may need a doctor rather than amplification. Someone whose two ears differ a lot may need a more individual assessment than a consumer self-fitting device provides.
The glasses are best understood as a new form factor for a category of hearing assistance that already exists — not a new category, and not a diagnosis.
The genuinely interesting part is not that Meta turned ordinary glasses into medical equipment overnight. It is that hearing-aid software can now use the microphones, processors and speakers already built into a pair of smart glasses to build a personalised amplification system without putting anything inside the ear.
That makes the feature real.
It just does not make it right for everyone.