Online Blind ABX Test – FLAC or MP3?

Think you can actually hear the difference between lossless FLAC and compressed MP3? Put your ears -and your audio gear- to the ultimate test. Our free, web-based blind ABX test challenges you to identify true high-fidelity audio without any visual bias. Grab your best headphones, trust your hearing, and find out if you’re a true audiophile or if 320kbps is all you really need.

Can you really hear the difference between a lossless file and a high-bitrate MP3? This blind ABX test lets you load your own two tracks, matches their level and challenges you to identify a hidden sample — then tells you, with a statistical p-value, whether your result beats pure chance. Nothing is uploaded.

How the ABX test works

You get reference A, reference B and a hidden X that randomly equals A or B. You decide which one X is. Repeat across several trials; if you get enough correct, the odds of guessing are tiny and the difference is audible to you.

If you suspect one of the two files is not what it claims to be, run it through the fake lossless detector first: it reads the spectrum and tells you whether it really came from a lossless source.

Frequently asked questions

What is an ABX test?

You hear reference A and reference B, then a hidden sample X that is randomly equal to A or B. You decide which one X matches. Repeating it several times and counting hits tells you, statistically, whether you can really hear a difference.

Are my files uploaded anywhere?

No. The files are decoded and played entirely in your browser with the Web Audio API. Nothing is sent to any server, which is also why it is fast and private.

Why does level matching matter?

Even tiny volume differences are heard as «quality» differences and ruin the test. The tool matches the perceived loudness of both tracks using K-weighting (the same idea as the LUFS standard, which boosts highs and filters subsonics to measure as the ear hears), so you compare the sound, not the level.

How many correct answers prove I hear the difference?

It depends on the number of trials. With 16 trials, 12+ correct gives p < 0.05 (less than 5% chance of luck). The tool computes the exact binomial p-value for you.

A blind, level-matched ABX test shows whether a difference is audible to you on your gear — not whether two formats are bit-identical.