Here is the short version, before you take anything: you are almost certainly not tone deaf. Real tone deafness — congenital amusia — affects somewhere around 4% of people, and it means the brain cannot reliably register pitch differences of about a semitone or more. That is an enormous gap. It is the distance between two adjacent keys on a piano. If you can hear that two notes are different when someone plays them side by side, you do not have it.
What most people mean when they say "I'm tone deaf" is that they sing badly, and singing badly is usually a motor problem rather than a hearing problem. This test separates the two. It measures the only thing that actually defines tone deafness: how small a pitch difference your ear can resolve.
You will hear two tones, one after the other. Say whether the second
tone was higher or lower than the first. There are ten trials, and the difference gets
smaller each time — starting at 20 cents and ending at 4 cents, where one cent is a
hundredth of a semitone.
Desktop: press P for higher, L for lower.
Mobile: tap the top half of the screen for higher, the bottom half for lower.
Headphones strongly recommended, and a quiet room. Bluetooth is fine for this
particular test — latency does not matter here, only pitch does.
This test is deliberately harder than a tone deafness screen needs to be. Every trial, including the easiest one, is finer than the threshold that defines amusia. That has a useful consequence and an honest limitation, and it is worth being clear about both.
What it can prove: if you correctly identified the 20-cent and 15-cent trials, you resolved pitch differences roughly five to seven times finer than the semitone-scale gap that people with amusia struggle with. That rules tone deafness out. Not "probably" — it rules it out.
What it cannot do: diagnose amusia. A poor score here means you struggled with differences far below the clinical threshold, which is common, unremarkable, and not evidence of a disorder. It also happens routinely with laptop speakers, background noise, or a distracted first run. If you genuinely cannot hear a difference when someone plays two adjacent piano keys, that is worth following up properly with the Montreal Battery of Evaluation of Amusia, which is the research instrument built for it.
One more caveat that most online tone deaf tests skip: with two answer choices, guessing gets you 50% by accident. Five correct out of ten is what pure chance looks like. Read your score against that baseline, not against zero.
Congenital amusia is a lifelong condition, present from birth, in which pitch processing develops abnormally despite normal hearing, normal intelligence, and normal exposure to music. People who have it typically cannot detect when a familiar tune has been altered, often cannot tell whether a note went up or down, and frequently describe music as sounding like noise or clatter rather than as unpleasant.
It is not a hearing loss. Audiometric thresholds in people with amusia are usually completely normal — they hear quiet sounds just as well as anyone. The difference sits further along the chain, in how the auditory cortex and frontal regions encode and compare pitch. There is also a heritable component: amusia runs in families at rates well above chance.
The distinction that matters for almost everyone reading this page is between perception and production. Amusia is a perception disorder. Bad singing is nearly always a production problem. They are unrelated, and the second one is common while the first one is rare.
Singing in tune is a three-stage act. You have to hear the target pitch, then contract a set of small laryngeal muscles precisely enough to produce it, then compare what came out against what you intended and correct mid-note. Most people who sing off-key are doing the first stage perfectly well and failing at the second and third.
The tell is obvious once you know to look for it: people who sing badly can almost always hear that they sang badly. That reaction is proof of intact pitch perception. A person with genuine amusia would not wince, because they would not detect the error.
The practical upshot is encouraging. Motor skills respond to practice in a way that perceptual deficits do not. Most self-described tone deaf adults get noticeably better at matching pitch within a few weeks of deliberate practice — the ear training guide lays out how.
Ten trials, each a two-alternative forced choice — the same basic design used by research-grade pitch discrimination tasks. A tone plays, then a second tone one second later, and you report the direction of the change. The base frequency is randomised between 150 Hz and 800 Hz on every trial so you cannot anchor on a remembered pitch, and the direction is randomised too.
The differences descend through 20, 15, 12, 10, 8, 6, 6, 5, 4 and 4 cents. For scale: a semitone is 100 cents, and most listeners start losing reliability somewhere between 10 and 25 cents depending on training and register. Trained musicians and audio engineers can often work down to 5 cents. Below about 4 cents, almost nobody is reliable, which is why the test stops there rather than pretending otherwise.
This is the same pitch engine used as one of the five sections in the full MusIQ music IQ test. If you want to see how it is scored inside that battery, the pitch section breakdown has the formula.
Take a pitch discrimination test like the one above. If you can reliably hear differences far smaller than a semitone, you are not tone deaf. The everyday sense of the phrase — "I can't sing" — is a different thing entirely and is not what the clinical term means.
The figure usually quoted is about 4%, and several researchers think the real number is lower once you exclude people who were tested badly or were not paying attention. Either way, the odds are strongly against you being one of them.
No, not in the clinical sense — impaired pitch perception is the condition. If you hear the differences, you do not have amusia, no matter how your singing sounds.
Congenital amusia is a lifelong neurological difference and is not currently reversible through training. The far more common problem — normal hearing plus an untrained voice — improves quickly, because motor control is trainable in a way that perception is not.
No. A hearing test measures the quietest sound you can detect. This measures the smallest pitch difference you can detect between two sounds you can hear perfectly well. The two are independent — you can have excellent hearing and mediocre pitch discrimination, or the reverse.
Even genuine amusia does not rule it out — rhythm perception is typically unaffected, and plenty of instruments reward timing and touch heavily. And since you almost certainly do not have amusia, the honest answer for nearly everyone is that nothing about your hearing is standing in the way.