Is this hearing test accurate?

It is a screening estimate, not a diagnosis. Without a calibrated audiometer and sound-treated booth, absolute decibel levels cannot be guaranteed. Use it as a starting point for a professional evaluation, not a substitute for one.

Do I need special equipment?

Closed-back, over-ear headphones in a quiet room. Earbuds will fit poorly and produce inconsistent results. The app guides you through a one-minute calibration step before testing.

Yes. All audio is generated and analysed on your device using the Web Audio API. Test results are stored only in your browser's localStorage. Nothing is sent to any server. There are no accounts and no tracking.

What frequencies does the test cover?

The full assessment tests 250, 500, 1000, 2000, 3000, 4000, 6000, and 8000 Hz in each ear independently. The quick screening covers 500, 1000, 2000, and 4000 Hz.

When should I see an audiologist?

If your pure-tone average is worse than 25 dB HL, if there is a notable difference between your two ears, if you experience sudden changes in hearing, one-sided tinnitus, ear pain or fullness, persistent dizziness, or difficulty understanding speech in noise.

An at-home audiogram for the hearing-curious

Listen carefully.

A pure-tone threshold test, a tinnitus matcher, an ambient-noise meter, and a longitudinal record of your hearing — calibrated against your own headphones, kept private on your device.

Not a medical device. A screening tool to inform conversations with a licensed audiologist.

Latest assessment —

No assessments yet.

Your audiograms and trends will appear here once you complete your first test.

Quick actions

Trend — pure-tone average

last 12 tests

Right ear Left ear ▼ lower is better

§ 01 · Calibration

A reference, set by you.

Consumer headphones are not laboratory transducers. Before each session we ask you to anchor a single reference tone to a level that just-barely-but-clearly sounds like quiet conversation. From this anchor we estimate hearing levels in decibels relative to typical normal hearing (dB HL).

Step 1 of 3

Headphone check

Wear your headphones the right way around. We will play short tones in each ear so you can verify left and right are correctly placed.

I heard each tone in the correct ear.

Step 2 of 3

Find a quiet room

No fans, no music, no conversation. The quieter your environment, the more accurate the threshold.

—

Step 3 of 3

Anchor the reference tone

A pulsed 1 kHz tone will play in both ears. Adjust your system volume until the tone sounds like quiet conversation, one room away — present, but unobtrusive. Then confirm.

Fine trim (within-app) 0.0 dB

Pure-tone threshold

Right ear

— / —

1000 Hz mid frequencies

— preparing —

A tone may or may not play. Press only when you hear one — even faintly.

Before we begin

A few conditions, briefly.

iThe room must be quiet. Close windows and doors.
iiWear over-ear headphones. Earbuds will fit poorly.
iiiCalibration must be current. .
ivAvoid testing right after loud-noise exposure (give your ears 12 hours).
vPress the response button at the faintest perception, not when comfortable.

Assessment results

Your audiogram

—

○ Right ear (air conduction)

× Left ear (air conduction)

↓ louder · ↑ quieter

Interpretation

PTA · Right — —

PTA · Left — —

Asymmetry —

Plain language

—

Suggested next steps

Severity bands

−10 to 25Normal

26 — 40Mild loss

41 — 55Moderate

56 — 70Mod. severe

71 — 90Severe

91 +Profound

Longitudinal record

History

—

Date	Mode	PTA R	PTA L	Class R	Class L

Per-frequency trend

Track each frequency over time. A widening gap on the right side of the chart often signals early noise-induced loss.

§ 04 · Tinnitus

Match the ringing.

Sweep through pure tones and narrowband noise to find the pitch and loudness that most closely match what you hear. Match values, when tracked over time, can help an audiologist tailor sound therapy.

Stimulus

Frequency 6000 Hz

Loudness −30 dB

Side

§ 05 · Sound exposure

Ambient noise meter.

Uses your microphone to estimate sound pressure level. Approximate, uncalibrated, and entirely private — audio never leaves the page. Useful for spotting relatively loud environments and tracking your daily noise dose.

— dB SPL · approx

Peak—

Avg (Leq)—

Duration00:00

Daily dose0%

Reference

A short field guide.

How the test works

We use a modified ascending–descending threshold method, presenting pure tones at calibrated frequencies — 250, 500, 1000, 2000, 3000, 4000, 6000, 8000 Hz — to each ear independently. Tones are randomised in timing to discourage anticipation. Your threshold at each frequency is the lowest level you reliably detect.

Pure-tone average (PTA) is the mean of thresholds at 500, 1000, and 2000 Hz — the speech-critical band. PTA underpins the loss-severity classifications used worldwide.

Reading an audiogram

Frequency runs across the X-axis on a log scale (low pitches left, high pitches right). Hearing level runs down the Y-axis from 0 dB HL at the top to 110+ at the bottom — so a steeper drop on the chart means worse hearing. Right-ear thresholds are marked with red circles; left-ear thresholds with blue crosses.

A "ski-slope" pattern — normal hearing in the lows, sharply diminished in the highs — is the classic signature of noise-induced loss, often beginning at 4 kHz.

When to see a professional

Any sudden change in hearing
One-sided hearing loss or tinnitus
PTA worse than 25 dB HL on this screening
Difficulty understanding speech in noise despite normal pure-tone thresholds
Pain, fullness, or persistent dizziness
Family history of progressive hearing loss

Protecting what you have

The 60/60 rule — under 60% volume, under 60 minutes at a time
Custom-moulded musician's earplugs preserve frequency balance
Step out of loud environments every 30 minutes for a quiet minute
Avoid testing your headphones with someone else's volume settings
Hearing loss is cumulative; days of recovery do not erase hours of damage

Limitations of at-home testing

This tool produces a screening estimate, not a diagnosis. Without a calibrated audiometer, sound-treated booth, and bone-conduction transducer, we cannot distinguish conductive loss (middle-ear) from sensorineural loss (cochlear/neural), cannot measure tympanic compliance, and cannot rule out retrocochlear pathology. Treat all results as a starting point for a professional evaluation, not a substitute for one.