Prescription Hearing Aid Lab Data: What Independent Testing Shows

What This Page Is, and What It Isn't

Prescription aids get measured the same way OTC ones do: five scores out of 5 (speech in quiet, speech in noise, own voice, music streaming and feedback), rolled into one SoundScore and a letter grade, plus a percentile badge on the product page. Everything here was read off HearAdvisor's own product pages on October 5 and 6, 2026.

Initial Fit vs Tuned Fit

HearAdvisor measures every device twice. The initial fit is the aid as it comes out of the box, at the manufacturer's default settings. The tuned fit is after a professional adjusts it to hearing-loss targets (HearAdvisor's EON Sphere page says NAL-NL2 targets at 55, 65 and 75 dB SPL) for the lab's test audiogram.

That split matters far more for prescription aids than for OTC ones. Four of the five OTC devices I checked score exactly the same at both fits. None of these ten does. If an aid is only as good as the person who programs it, this is where that shows up in numbers.

The Scores

Thirteen prescription aids, highest SoundScore first. Each cell shows initial fit → tuned fit. One number means both fits matched.

What Tuning Changed

The biggest move is in noise on the two Phonak Sphere aids. Speech in noise goes from 3.7 to 4.5 on the EON Sphere and from 3.1 to 4.5 on the Sphere Infinio Ultra. The Signia Silk, an invisible-in-ear aid, rises from 1.4 to 2.2 in noise and from 2.7 to 4.6 in quiet. Elsewhere it's a few tenths up (Zeal 2.7 to 3.0, Signia MaX 2.1 to 2.8, Horizon Go IX 2.3 to 2.6, Omega AI 2.7 to 2.9), flat (the Costco Rexton, 2.5), or down (Widex Allure 3.0 to 2.3, Intent 3.5 to 3.1, Real 3.0 to 2.9).

Tuning isn't a free upgrade. Music streaming falls after tuning on eight of the twelve aids that stream (Real 4.2 to 2.8, Horizon Go IX 4.1 to 2.7, Reveal 1 4.2 to 3.5), stays put on two, and rises on two. Intent, Real and the Widex Allure score lower on speech in quiet after tuning, and the Allure's feedback score drops from 3.8 to 1.7. I can't tell you why from these numbers, and the lab's pages don't say. What they do show is that "professionally tuned" and "better on every measure" are different things.

Own voice is the weak spot across the set. It runs from 1.3 on the Intent and 1.8 on the Horizon Go IX up to 3.7 on the Signia MaX and 3.6 on the Rexton. It also moves with tuning on the Silk, from 3.6 down to 1.5. HearAdvisor's own-voice score is an occlusion measure, how the aid sounds to its own wearer's voice, and not a judgment of how natural your voice sounds. It is the same pattern as the OTC devices, where the best own-voice score was 2.9.

The Costco Rexton

What the extra money buys is mostly noise performance. The Rexton scores 2.5 in noise at both fits, against 4.5 tuned on the two Phonak Sphere aids. It also has one of the best own-voice scores here (3.6) and a top-27% badge against top-1% and top-5% for the Phonaks. So the gap between $1,600 and $4,600 is real in a noisy restaurant, and much smaller in a quiet room. Whether it's worth it depends on where you struggle. That's your call, not mine.

hear.com's Horizon Go IX

The Horizon Go IX scored 4.4, Grade A, top 44%. It's the mid-pack aid in this set, with one of the lowest own-voice scores in the set (1.8) and a big drop in music streaming after tuning (4.1 to 2.7). HearAdvisor lists it at $6,500. That doesn't match hear.com's own price page, so check hear.com for what you would actually pay. My hear.com review has the tier prices I read there.

What the Lab Hasn't Tested

Some aids you can buy today have no score here. As of October 6 I found no HearAdvisor product page for these: Starkey Omega AI+, Starkey Signature Series, ReSound Sensia, Phonak Virto Infinio and Phonak Naída Lumity. That last family is mine. (HearAdvisor does list a Phonak Audéo Lumity, which is a different aid.)

Newer models don't inherit an older model's score. The EON Sphere and the Sphere Infinio Ultra are different aids with different scores (5.0 and 4.9), and so are Oticon's Intent (2024) and its Reveal 1 (2026). When a retailer or a brand says the new aid is "the same platform, only better," that's a claim, not a measurement.

How to Use This

• 1
  Compare the noise score first if restaurants are your problem. It's the measure that varies most between these aids, and the one a good fitting moves most.
• 2
  Ask the provider how they tune. The scores above come from fitting to targets and measuring the result. Ask whether your provider does real ear measurements, because a fitting that skips them isn't the one HearAdvisor tested.
• 3
  Treat the lab as one input. It tests a mild-to-moderate loss on a standard audiogram. It can't measure your provider, your ear canal, your follow-up visits, or what the aid costs you after year one.

Source and Methodology Note

Cite or Share This Data

Journalists, researchers, clinicians and bloggers are welcome to cite the analysis here. Please link back to this page and credit HearAdvisor as the original testing source. For any individual aid's scores, cite HearAdvisor's own product page.

Keath D. (2026). Prescription hearing aid lab data: What independent testing shows. HearLifeRestored. https://hearliferestored.com/prescription-hearing-aid-lab-data

Also see the OTC hearing aid lab data and the OTC vs prescription comparison.