How to A/B Compare Audio Like a Pro — Train Your Ears

"It Sounds Better" Is Not Good Enough#

You applied an EQ, ran a compressor, or used a mastering preset — and it "sounds better." But does it really? Or did your ears just trick you?

Human hearing has a well-documented flaw: louder sounds are perceived as better. This is called loudness bias. If mastering raises the overall level even slightly, your brain will prefer the processed version — even if the actual tonal quality got worse.

A/B comparison exists to eliminate this illusion and let you make decisions based on real sonic differences.

What Is A/B Comparison?#

A/B comparison means switching between the unprocessed (A) and processed (B) versions of your audio to judge changes objectively.

Why It Matters#

Audio processing is full of perceptual traps:

Bias	What Happens	Example
Loudness bias	Louder = better	Post-mastering sounds "improved" but it's just louder
Confirmation bias	You hear what you expect	Expensive plugin must sound better, right?
Order effect	The second sample sounds better	A→B listening order favors B
Memory decay	You forget the first sound quickly	After 5 seconds, accurate comparison is impossible

A/B comparison is the scientific approach to minimizing all of these.

How to A/B Compare Correctly#

Step 1: Match the Levels#

This is the most critical step. If the before and after versions differ in loudness (LUFS), loudness bias invalidates your comparison.

• Normalize both versions to the same LUFS value
• A difference within 0.5 LUFS is acceptable
• If the gap exceeds 1 LUFS, the comparison is unreliable

Step 2: Switch Instantly#

Toggle between A and B as fast as possible — ideally within 0.5 seconds. Human echoic memory degrades rapidly, and after 5 seconds you can no longer compare accurately.

Step 3: Repeat the Same Section#

Don't compare different parts of a song. Loop the same section and listen at least 3 times, focusing on specific elements each pass.

Step 4: Go Blind#

If possible, compare without knowing which is A and which is B. This eliminates confirmation bias. Have a friend toggle randomly, or use a tool with blind test functionality.

What to Listen For#

Don't listen to the "overall vibe." Instead, focus on specific elements each time.

Frequency Band Checklist#

Band	Range	What to Check
Low	20–200 Hz	Kick presence, bass clarity, boominess
Low-mid	200–500 Hz	Warmth, muddiness, vocal body
Mid	500 Hz–2 kHz	Vocal clarity, instrument separation
Upper-mid	2–8 kHz	Presence, vocal intelligibility, harshness
High	8–20 kHz	Air, cymbal shimmer, sibilance

Dynamics Check#

• Transients — Are kick and snare attacks preserved?
• Attack sharpness — Has the edge been lost?
• Sustain/decay — Is the reverb tail natural?
• Overall dynamics — Is the contrast between loud and quiet sections maintained?

Stereo Image Check#

• Width — Has the horizontal spread changed?
• Panning — Are instruments still in the right positions?
• Mono compatibility — Does it collapse when summed to mono?

How to Run a Blind Test#

DIY Method#

1. Load both files into a player

Look away and randomly press play on either version 3. Note which one sounds "better" 4. Repeat 10 times and check consistency

If you pick the same version 7+ times out of 10, the difference is audible. A 5:5 or 6:4 split means there's no meaningful difference.

Common Discoveries#

Blind testing often reveals surprising truths:

• Expensive plugins and free ones sound the same — more common than you think
• The unprocessed version sounds better — over-processing backfires
• Differences exist in only one frequency range — not everything changes

DeckReady's A/B Preview Feature#

DeckReady includes a built-in A/B preview function for comparing audio before and after processing.

Feature	Description
Instant switching	Toggle between processed and unprocessed with zero latency
Auto level matching	LUFS values are automatically aligned to eliminate loudness bias
Position lock	Playback position stays synced during switching
Waveform display	Visual comparison of before and after waveforms

The auto level matching is particularly important. Most A/B tools don't correct for volume differences, leaving you vulnerable to loudness bias. DeckReady automatically matches LUFS values so you're evaluating pure tonal changes.

Use Cases#

• Preset selection — Compare Club, Lounge, and Broadcast presets to find the best fit
• Custom tweaks — Adjust LUFS targets or EQ settings with real-time A/B feedback
• Processing necessity — Objectively decide if processing is even needed

Ear Training Exercises#

A/B comparison skills improve with regular practice.

Recommended Exercises#

1. EQ training — Boost a specific band by ±3 dB and identify which frequency changed
2. Compression training — Identify light, medium, and heavy compression by ear 3. Format comparison — Blind test WAV vs. MP3 at 320 kbps and 128 kbps 4. Active listening — During your commute, consciously focus on individual frequency bands

Skill Benchmarks#

Level	What You Can Hear
Beginner	3 dB+ EQ changes, WAV vs. MP3 128 kbps
Intermediate	1–2 dB EQ changes, lossless vs. MP3 320 kbps
Advanced	0.5 dB changes, subtle compression differences

Summary#

A/B comparison transforms audio judgments from gut feeling to evidence-based decisions.

1. Match levels — The single most important step to eliminate loudness bias
2. Switch instantly — 0.5 seconds or less 3. Listen by frequency band — Focus on specifics, not the overall impression 4. Use blind testing — Remove confirmation bias 5. Use DeckReady's A/B feature — Level-matched comparison with zero effort

Building a reliable internal reference for "good sound" is a lifelong asset for any DJ or music producer.