Audio Blind Test Comparator
Verify if you can actually hear a difference between 2 files.Load files A/B > Start Blind Test > Guess multiple times which file is playing > Reveal Results.>
Load both files above to begin. Each trial randomly assigns A or B as the mystery file X. Listen freely, then vote. Results and statistics reveal only at the end — no bias during the test.
Frequently Asked
How do I use this ABx Test?
- Load your two files — click slot A and slot B, or drag and drop audio files directly onto each slot. Any browser-supported format works: WAV, FLAC, MP3, AAC, OGG, and more.
- Preview and compare freely — use the Player section to listen to A and B back to back. The timeline scrubs both files; switching slots carries your playback position forward so you’re always comparing the same moment.
- Start the Blind Test — click the “Start Blind Test” button.
Each trial plays a mystery file X (secretly either A or B). Listen to X, then use the A and B buttons to compare, then click A or B to cast your vote. - Repeat for all 16 trials — no right/wrong feedback is shown during the test to keep it fully unbiased. You can stop before completing the 16 trials but we suggest doing at least 10, otherwise your results may not be accurate.
- Review your results — accuracy percentage, p-value, per-trial breakdown, and a plain-language statistical interpretation are revealed at the end.
Tip: Unless if your studio, use headphones in a quiet environment. Listener fatigue is real — take a short break before starting the test if you’ve been mixing for hours!
Key features
Gain Match is enabled by default and applies RMS-based gain compensation in real-time between the two files. This ensures that volume differences — one of the most common sources of bias in audio comparisons — don’t influence your judgment. The dB offset applied to each file is displayed as a badge. For the most rigorous results, consider LUFS-normalizing your files beforehand in your DAW (e.g. with a loudness normalization plugin or export option), then disabling Gain Match to confirm the compensation is minimal.
Easy ABx Test in your browser: Simply click the “Start Blind Test” button after loading files A and B and guess which file you think is playing a few times before accessing the results.
ABx test allows you to tell via a less-biased test procedure if you can actually hear a difference between two files.
Statistical reporting — results include a one-tailed binomial p-value, which is the same statistical method used in peer-reviewed audio research and professional ABX software. A p-value below 0.05 is the standard threshold for claiming audible discrimination.
Drag and drop is supported on both file slots for fast workflow integration.
Are my audio files uploaded anywhere? How is data handled?
No — nothing ever leaves your device. This tool runs entirely inside your browser using the Web Audio API. Your audio files are decoded locally in memory, never written to disk by the tool, never sent to a server, and never logged or stored anywhere.
Suggestions for best results
- Gain-match carefully before hand. Even with Gain Match ON, RMS compensation is a broadband average. If your files have very different spectral or dynamic characteristics (e.g. one is heavily limited, one is unmastered), consider using an integrated LUFS meter in your DAW for pre-matching. Or something like the -neat- free plugin Ghz Loudness by Goodhertz, which let’s automatically gain match based on short term LUFs!
- Run multiple sessions. A single 16-trial test has limited statistical power. Running the test twice on different days and combining results gives a much more reliable picture.