THE AUDIO
GLOSSARY.

Stem Mastering

Stem mastering masters a track from a small number of instrument groups (stems) — typically drums, bass, music, and vocals — instead of a single stereo file. The extra separation lets the engineer rebalance elements and solve problems that can't be fixed on a summed stereo mix, making it a middle ground between stereo mastering and a full mix.

Headroom

Headroom is the gap between the loudest peak in your mix and 0 dBFS (digital full scale). Leaving 3-6 dB of headroom on your mix bus before mastering gives the mastering chain room to process — EQ, compression, and limiting all change level, and a mix that's already slammed against 0 dBFS leaves nowhere to work.

Dithering

Dithering adds a tiny, carefully-shaped layer of noise when reducing bit depth (for example 24-bit to 16-bit for CD or some distributors). It randomizes quantization errors so quiet fades and reverb tails stay smooth instead of developing gritty digital distortion. Apply it once, as the very last step, only when actually reducing bit depth.

LUFS

LUFS (Loudness Units relative to Full Scale) measures how loud a track actually sounds to human ears, not just its peak level. Streaming platforms normalize to a LUFS target — Spotify and YouTube to about -14 LUFS — so mastering to the right LUFS is how you control loudness after normalization.

True Peak

True peak (dBTP) measures the actual analog peak a signal reaches after digital-to-analog conversion, including inter-sample peaks that sit between samples. It can run 1-3 dB higher than the sample peak your DAW shows. Master to a -1 dBTP ceiling so streaming codecs and consumer DACs don't clip.

Limiter

A limiter is a fast compressor with a very high ratio that stops a signal from exceeding a set ceiling. In mastering, a brickwall limiter is the final stage — it raises overall loudness while catching peaks so nothing clips. Push it too hard and you crush transients and add distortion, so it's about finding the loudest clean setting, not the loudest possible.

Dynamic Range

Dynamic range is the difference between the quietest and loudest parts of a track. Wide dynamic range keeps transients punchy and lets a chorus hit harder than a verse; narrow (heavily compressed) range sounds consistently loud but can feel flat and fatiguing. Good mastering controls dynamic range without destroying it.

Multiband Compression

Multiband compression splits the signal into frequency bands (for example lows, mids, highs) and compresses each independently. It lets you tame a boomy low end or harsh highs without pumping the whole mix, making it a precise tool for mastering and for controlling problem frequencies that a full-band compressor would over-react to.

Sidechain Compression

Sidechain compression uses one signal to trigger compression on another. The classic use: the kick drum ducks the bass (or the whole mix) every time it hits, so the kick punches through cleanly. It's both a mixing fix for kick-bass clashes and a creative pumping effect central to house, EDM, and modern pop.

Saturation

Saturation adds subtle harmonic distortion that emulates analog gear — tape, tubes, transformers, consoles. It makes sounds feel warmer, denser, and louder without raising the peak level, because it adds harmonics rather than gain. Used gently across a mix or master, it's the difference between a clean-but-sterile record and one with analog character and glue.

Mid/Side (M/S) Processing

Mid/side processing splits a stereo signal into a mid channel (what's common to both speakers — usually vocals, kick, bass, snare) and a side channel (what differs — usually reverb, wide synths, stereo effects). Processing them independently lets you EQ, compress, or widen the center and the sides separately for precise stereo control.

Stereo Imaging

Stereo imaging is the perceived width and placement of sounds across the left-right field. Good imaging makes a mix feel wide and three-dimensional while keeping the important elements — vocal, kick, bass, snare — anchored in the center. Over-widening can sound impressive on headphones but collapse or go hollow in mono.

Gain Staging

Gain staging is setting the level at each point in your signal chain so every plugin and channel receives an optimal input — not too hot, not too quiet. Proper gain staging keeps analog-modeled plugins in their sweet spot, prevents cumulative clipping, and gives you consistent headroom from the first insert to the master bus.

De-esser

A de-esser is a frequency-specific compressor that tames harsh sibilance — the 's', 't', 'sh', and 'f' sounds in vocals that spike in the 5-10 kHz range. It reacts only when those frequencies get too loud, smoothing them without dulling the rest of the vocal. Uncontrolled sibilance is one of the fastest tells of an amateur vocal mix.