ABX Blind Listening Test – Level, Pitch, Frequency, Timing & Bit Depth
ABX comparison0 / 0
- Below 5 %: you hear it
- 5–20 %: maybe, test more
- Above 20 %: likely guessing
- Use headphones in a quiet room
- Never raise volume to discomfort
- Screening only, not diagnosis
- Runs inside your browser
- No upload, no account
- Nothing saved after you close
ABX blind tests for hearing differences
In an ABX test you hear A and B, then an unknown X, and say which one X matches. Repeat ten times; the statistics tell you whether you really hear a difference.
Why blind testing
Expectation changes what we hear.
Level and pitch tests
Most listeners hear 1 dB level steps on tones and around 5–10 cents of pitch at 440 Hz; trained musicians do better.
Timing tests
Two clicks a few milliseconds apart often fuse into one; the threshold for hearing them as two is around 2–5 ms.
How to Use the
Blind Listening Test
- 01
Choose
select a test from the list.
- 02
Learn
play A and B a few times.
- 03
Identify
play X and answer whether it is A or B.
- 04
Repeat
do at least 10 trials and read the guess probability.
Why blind testing
Expectation changes what we hear. Blind tests remove labels so only audible differences remain.
Level and pitch tests
Most listeners hear 1 dB level steps on tones and around 5–10 cents of pitch at 440 Hz; trained musicians do better.
Timing tests
Two clicks a few milliseconds apart often fuse into one; the threshold for hearing them as two is around 2–5 ms.
Bit depth and absolute polarity
At normal volume, 16-bit audio is transparent for most people. Absolute polarity is very hard to hear on music; a kick drum is the best-case signal.
Absolute pitch
Absolute pitch is the ability to name a note without reference. Try the C vs C# test after a break to avoid relative cues.
About blind testing
Purpose
To find out whether you can really hear a difference rather than expect one.
What is being compared
Each test pairs two signals that differ in one property: level, pitch, bandwidth, timing, noise, resolution or polarity.
Take the challenge
Aim for 12 correct out of 16; less than 5 % probability of guessing is strong evidence.
Specific tests
Blind timing test
Clicks separated by a few milliseconds test temporal resolution.
Upper hearing limit blind test
The frequency-limit options compare a high tone with silence so you cannot guess.
How to use blind tests for engineering decisions
Blind ABX tests show whether a difference is audible to you, not just expected. Typical targets: a level difference of about 1 dB is the smallest most people reliably detect; timing differences of a few milliseconds become audible as flams; pitch differences of a few cents are heard by trained listeners. Score at least 12 of 16 trials before trusting a result.
Related guides and fixes
Learn more in these guides: Hearing frequency test explained, Hearing health guide, Decibel levels chart, Safe noise levels.
Frequently Asked Questions
Q.How many trials do I need?
At least 10; 16 or more gives stronger evidence.
Q.Is X random?
Yes, a new random choice after each answer.
Q.Can headphones affect results?
Yes, use the same setup each time.