What Is Dynamic Range Compression?

The word 'compression' causes constant confusion in audio because it means two completely different things. Dynamic range compression (the topic of this guide) is a real-time signal processing technique used in mixing and mastering to control loudness dynamics. It is applied by a compressor — a hardware device or software plugin. File compression (used in formats like MP3, FLAC, and OGG) is a data encoding technique that reduces storage size. An MP3 uses psychoacoustic file compression. A vocal recording uses dynamic compression. These are unrelated processes. When a producer says 'compress that vocal,' they mean dynamic range control. When a developer says 'compress the audio,' they often mean format encoding. Context matters.

A dynamic range compressor automatically reduces the level of an audio signal when it exceeds a set threshold. Before the threshold, the signal passes through unchanged. Above the threshold, the compressor attenuates the signal by a ratio you define. A 4:1 ratio means that for every 4 dB the signal exceeds the threshold, only 1 dB gets through. A 10:1 or higher ratio is called limiting — essentially a hard ceiling on the signal level. The result is a more consistent level: loud parts get quieter, the overall dynamic range is reduced. You then often apply makeup gain to bring the compressed signal back to the original perceived volume, resulting in audio that is louder and more consistent.

Threshold sets the level at which compression begins — typically measured in dB. Signals below the threshold are unaffected. Ratio defines how aggressively the compressor reduces gain above the threshold. Attack controls how quickly the compressor engages after the signal crosses the threshold — a fast attack catches transients immediately, a slow attack lets the initial hit through. Release controls how quickly the compressor returns to unity gain after the signal drops below the threshold. Knee describes the transition between uncompressed and compressed states: a hard knee engages compression abruptly at the threshold, a soft knee gradually increases compression as the signal approaches and crosses the threshold. Each parameter shapes the character of the compression differently and interacts with the sonic nature of the material being compressed.

On a vocal track, compression smooths out the level differences between whispered phrases and belted notes, making the whole performance sit consistently in the mix without constant manual volume automation. On drums, a fast attack and fast release on a snare compressor tightens the transient response and adds punch. A slower attack on a kick drum let the initial transient through before clamping down, enhancing the click of the beater. Bus compression (applied to groups of instruments) glues elements together — the slight level interaction between kick, bass, and other low-end elements creates a sense of cohesion. Parallel compression (mixing compressed and uncompressed signals) is used on drums to preserve transient energy while adding sustain and weight.

In mastering, compression levels the playing field across an album and controls the overall dynamic character of the music. Heavy limiting raises the integrated loudness (measured in LUFS) to competitive streaming levels. Spotify normalizes tracks to -14 LUFS. Apple Music normalizes to -16 LUFS. YouTube Music to -13 to -14 LUFS. Over-compression to chase high LUFS creates what audiophiles call the loudness war artifact — crushed transients, fatiguing listening experience, and a flat, airless sound. Modern best practice is to master to the streaming platform targets, accepting that the normalization algorithm will bring every track to the same perceived loudness regardless of how loud you master. The optimal strategy is to maximize dynamic range within the target LUFS, not to maximize peak level.