Compression

Chapter 4 shaped which frequencies stand out at any given moment. This chapter deals with a different axis entirely: how consistently loud a recording stays over time. Compression evens out the gap between a recording's loudest and quietest moments — a technique used constantly in podcasting and voiceover work.

What Compression Actually Does

A raw voice recording naturally varies in loudness — a word spoken with emphasis is louder than one trailing off at the end of a sentence. Compression automatically turns down the loud moments once they cross a set level, narrowing the overall gap between the loudest and quietest parts of the recording. The result is a track that sounds more consistently present throughout, without a listener needing to constantly adjust their own volume.

The Core Controls: Threshold, Ratio, Attack, Release

Threshold sets the loudness level above which compression starts acting — anything below the threshold is left alone. Ratio sets how strongly the compressor turns down whatever crosses that threshold (a higher ratio means more aggressive turning-down). Attack sets how quickly the compressor reacts once the threshold is crossed, and release sets how quickly it lets go once the signal drops back below it. These four controls together shape both how much compression happens and how naturally it's felt.

Why Podcasting and Voiceover Rely on Compression So Heavily

Spoken-word content is especially prone to big loudness swings — a host leaning toward the microphone, a guest speaking more quietly than the host, natural emphasis on certain words. Compression is what keeps a podcast episode listenable in a car, on headphones, or through a phone speaker, without a listener needing to ride the volume knob themselves. This is exactly why it's treated as a near-constant technique in this kind of work, rather than an occasional creative effect.

Makeup Gain: Restoring Volume After Compression

Since compression turns the loud parts down, the overall recording often ends up quieter than it started. Makeup gain raises the whole signal back up afterward, bringing the now-narrower dynamic range up to a comfortable overall level — similar in spirit to Chapter 3's normalization, but applied after compression has already narrowed the gap between loud and quiet.

Over-Compression: The "Squashed" Sound

Pushed too hard — a very low threshold combined with a very high ratio — compression flattens a recording's natural dynamics almost entirely, producing a "squashed," fatiguing sound where every word lands at nearly the same loudness with none of the natural rise and fall of real speech. A more moderate setting still narrows the loudness gap meaningfully while leaving some natural variation intact.

Hands-On Exercises