Free tool
Audiogram reader
Type in the numbers from your hearing test and this plots them, averages them, grades them on both published scales, and — the part that is usually left out — describes the shape of the curve. In a cohort of 51,989 patients, shape was what tracked with tinnitus, and the average that leads every report is precisely what throws the shape away.
Values are decibels hearing level, in 5 dB steps. Leave a frequency blank if it is not on your report.
O right ear · X left ear · air conduction only. Down is worse. Range shown is -10 to 120 dB HL.
Enter the thresholds from your report and the plot, the averages and the shape appear here. Nothing you type is sent anywhere — this runs entirely in your browser.
The criteria this tool uses
- Notch — worst threshold at 3, 4 or 6 kHz, at least 10 dB below the best of 500–2000 Hz and recovering at least 10 dB by 8 kHz.
- Sloping — high frequencies (4, 6, 8 kHz) average at least 15 dB worse than low (250, 500, 1 kHz); steeply at 30 dB. Rising is the same margin the other way.
- Asymmetric — the two ears differ by at least 15 dB at any single frequency.
Those numbers are stated because they had to be chosen. There is no standardised definition of an audiometric notch — a 2006 study in Ear and Hearing opens by saying so, and found that at least five of its six experts agreed on whether a notch was present in only 71% and 72% of cases, with depth and the reviewer’s own specialty both affecting the judgement. A tool that gave you a confident label without showing its criterion would be claiming a precision the literature says does not exist.
Why shape rather than severity
A 2020 study in Otology & Neurotology reviewed 51,989 adults who had an initial audiometric evaluation at Massachusetts Eye and Ear between 2000 and 2016, and asked which of them came in because of tinnitus. About 20% did. Tinnitus as the primary complaint was statistically associated with particular configurations: most prevalent for notched and steeply sloping losses, and relatively uncommon in adults with flat ones.
The authors read that as support for tinnitus arising from sharp discontinuities in surviving nerve supply and cochlear amplification — the edge of the damage rather than its size. Which is why a dip you could cover with a fingertip can matter more than a larger, evener loss.
Hold it as an association across a clinic population rather than a rule about a person. Plenty of people have flat losses and severe tinnitus. What the finding earns is a reason to look at the shape at all, which almost nobody is invited to do. How to read your audiogram covers the chart itself in full.
What this tool does not do
- Bone conduction. Only air conduction thresholds go in, so an air–bone gap — the finding that points to a mechanical, often treatable cause — is not something this can detect. That comparison needs the full report.
- Anything above 8000 Hz. Standard audiometry stops there, and so does this.
- Hearing in noise. A different question with a different test — see the digits-in-noise test.
- Your tinnitus. Nothing on an audiogram shows it. How tinnitus is measured covers the instruments that do.
- Diagnose. It arranges numbers you already have and shows you what published scales say about them. A 15 dB difference between your ears flagged here is a prompt to ask a question, not an answer to one.
Frequently asked questions
What do I type in?
The numbers from your audiogram, in decibels hearing level, for each frequency and each ear. On a printed report they are usually listed in a table beside the chart; if you only have the chart, read each O and X off against the dB scale on the left. Air conduction only — the O and X marks, not the bracket-shaped bone conduction symbols. Leave any frequency blank if it was not tested.
Is anything I enter sent anywhere?
No. Every calculation runs in your browser, nothing is transmitted, and nothing is stored — reloading the page clears it. There is no account, no upload and no analytics on what you type.
Why does the shape matter more than the average?
Because the average throws the shape away. The pure tone average is the mean of your thresholds at 500, 1000 and 2000 Hz, so someone with normal hearing there and a deep dip at 4000 Hz can have a completely normal average and a clearly abnormal audiogram. In a study of 51,989 patients, tinnitus as the primary reason for the visit was statistically associated with notched and steeply sloping configurations, and was relatively uncommon in flat losses — an association with shape, not with severity.
Where do the notch and slope criteria come from?
They are this tool's own stated conventions, shown on the page beside every result. That is deliberate: a 2006 study in Ear and Hearing opens by noting that no standardised definition of an audiometric notch exists, and found that at least five of six expert reviewers agreed on whether a notch was present in only 71% and 72% of cases. Since a criterion has to be chosen, this tool shows you which one it chose.
Can this tell me whether I have hearing loss?
It can tell you where your numbers fall on two published scales, which disagree with each other. The WHO defines hearing loss as thresholds worse than 20 dB and disabling hearing loss as worse than 35 dB in the better ear; the ASHA table calls 16 to 25 dB slight rather than normal. What it cannot do is diagnose anything, interpret bone conduction, or replace the person who ran the test.
Configuration findings from Lewis, Jahn et al., Otology & Neurotology, 2020. Asymmetry and hyperacusis from Jahn & Polley, Hearing Research, 2023. Expert agreement on notch identification from Rabinowitz et al., Ear and Hearing, 2006.
Degree-of-loss table from the American Speech-Language-Hearing Association; the 20 dB and 35 dB thresholds from the World Health Organization.