Auditory Masking

When one sound hides another sound so you simply cannot hear the quieter one.

A loud masker raises the hearing threshold (blue dome); the quiet maskee (red) falls under it and vanishes until EQ carves space.

What it is

A louder sound hides a quieter sound so your ear simply cannot detect the quiet one.

Key facts

Masker = the loud sound. Maskee = the quiet sound it hides.
Critical bands: the ear splits sound into ~24 frequency bands (Bark scale); masking is strongest WITHIN one band.
Simultaneous (frequency) masking: the masker raises the audibility threshold of nearby tones, mostly upward in pitch ('upward spread of masking').
Temporal masking: pre-masking lasts ~5-20 ms BEFORE the loud hit; post-masking lasts ~50-200 ms AFTER it.
+10 dB = roughly twice as loud (perceived). +6 dB = double sound pressure. +3 dB = double power.
Just-noticeable level step ~1 dB; a maskee 6-10 dB under a masker in the same band is effectively gone.
Speed of sound in air = 343 m/s at 20 C (about 1235 km/h).
Kick drum and bass guitar both sit ~40-120 Hz, so they mask each other hardest (same critical bands).
MP3 / AAC codecs use a psychoacoustic masking model to delete masked audio you cannot hear, saving up to ~90% data.
Equal-loudness (Fletcher-Munson): ear is most sensitive ~2-5 kHz, so masking there wrecks intelligibility most.

How it works

Two sounds overlap in frequency and time.
The louder one (masker) raises your ear's threshold in that critical band.
The quieter one (maskee) now sits below threshold = you hear nothing.
Closer in pitch plus louder masker = stronger masking.
Masking spreads more upward in frequency than downward.
Fix: separate the sounds in frequency (EQ) or time so neither buries the other.

Real examples

Kick vs bass guitar fighting in the 60-100 Hz mud.
Lead vocal vanishing the moment distorted guitars enter.
Hi-hat hidden under a loud crash cymbal (same high band).
Snare swallowed by a wall of synths in the 200-500 Hz region.
Announcer's voice lost under loud crowd noise at a venue.

How it helps in live sound

Carve a 2-4 dB cut at ~60-80 Hz on the bass to let the kick punch through.
High-pass non-bass channels (vocals ~100 Hz, guitars ~80 Hz) to clear low-end masking.
Boost vocal presence ~2-5 kHz where the ear is most sensitive to win the masking war.
Use sidechain/ducking so the kick momentarily dips the bass (temporal de-masking).
Pan competing instruments left/right so each ear gets a clearer maskee.
If a sound is buried, EQ-cut its masker before you reach for more volume.

Everyday analogy

It is like trying to hear your mate talk right next to a running blender: their voice is still there, but the noise swamps the exact ear band that would have caught it.

Watch out

Myth: turn the quiet sound UP to fix masking. Truth: that just raises the whole mix and re-buries it. Instead CUT the masker with EQ to open a frequency window.

Fun fact

MP3 only works because of masking: the encoder works out which sounds are masked and permanently deletes them, throwing away most of the data while you hear no difference.

Key takeaways

Loud sound raises your hearing threshold so the quiet one drops out of audibility.
Strongest when both sounds share the same critical band (close in pitch).
Masking spreads upward in frequency more than downward.
Temporal masking hides sounds ~10 ms before and up to ~200 ms after a loud hit.
EQ carving creates space so sounds stop masking each other.
Cut the masker, don't just boost the maskee.

← Previous

Equal-Loudness Contours

Temporal Masking

☰ All 123 concepts