What Is Multi‑band Compression?

Multi‑band compression is a dynamic processing technique that splits an audio signal into separate frequency regions, applies independent compression to each region, and then recombines them. Unlike single‑band compressors, which affect the entire frequency spectrum with one set of parameters, multi‑band compression gives you surgical control over specific parts of the mix. This makes it an indispensable tool for mastering engineers, mixing engineers, and producers who need to tame harsh resonances, tighten low‑end, or balance vocal presence without altering adjacent frequencies.

At its core, multi‑band compression is about frequency‑dependent dynamics. A typical multi‑band compressor contains between two and six bands, each with its own threshold, ratio, attack, release, and gain controls. The crossover points between bands are adjustable, allowing you to define exactly where one band ends and another begins. This level of control is what sets multi‑band compression apart from standard dynamic processing and makes it especially powerful for complex, dense mixes.

How Multi‑band Compression Works

The processing chain of a multi‑band compressor involves three fundamental stages:

  1. Frequency Splitting – The incoming audio passes through a set of crossover filters that divide the signal into independent frequency bands. The crossover frequencies are user‑selectable and can be set to match the specific problem areas in your mix.
  2. Independent Compression – Each band is routed to its own compressor block. You can set a different threshold, ratio, attack, release, and makeup gain for each band. This allows you to apply heavy compression to a sibilant high‑mid range while leaving the low end untouched.
  3. Recombination – The processed bands are summed back together to form a unified audio signal. Careful gain staging is required here to avoid phase cancellation or frequency cancellation at the crossover points.

Modern multi‑band compressors use linear‑phase or minimum‑phase crossovers to minimise artifacts. Linear‑phase crossovers preserve phase relationships but introduce latency, while minimum‑phase crossovers have less latency but can cause phase shift near the crossover frequencies. Choosing the right crossover type depends on whether you are mixing in real time or performing offline processing.

Crossover Design and Band Overlap

Crossover points are crucial. If bands overlap too much, the compressor may work against itself. If they are too far apart, you may create audible holes in the spectrum. A good rule of thumb is to set crossovers at frequencies where you know the mix’s energy naturally separates – for example, 80–100 Hz to split bass from low mids, 200–300 Hz to isolate boxiness, 2–3 kHz to control harshness, and 6–8 kHz for sibilance control. Many engineers also use a linear‑phase crossover for mastering to avoid transient smearing.

Attack and Release Considerations

Each band’s attack and release settings should be tailored to the material in that frequency range. Low‑frequency content (e.g., kick and bass) typically benefits from slower attack times (10–30 ms) to allow the initial transient through, while high frequencies (e.g., cymbals, vocal sibilance) often need fast attack times (1–5 ms) to catch sharp peaks. Release times should be set to match the decay characteristics of the source material. Overly long release times can cause pumping, while overly fast release can distort the envelope.

Single‑band vs. Multi‑band Compression

Standard (single‑band) compression works well when you need to even out overall dynamics without changing the tonal balance. But it has a significant limitation: a loud high‑frequency spike can trigger the compressor even if the low end is perfectly stable, causing an unwanted volume drop in the bass. Multi‑band compression solves this by processing each frequency region independently. For example, you can compress a vocal’s sibilant band without affecting its body, or tighten a kick drum without dulling the overheads.

Another advantage of multi‑band compression is the ability to apply different ratios and thresholds per band. In a dense rock mix, you might use a 4:1 ratio on the low mids to control muddiness, a 2:1 ratio on the midrange to maintain punch, and a 6:1 ratio on the highs to tame harshness. Single‑band compression cannot achieve this without external dynamic EQ or parallel processing chains.

Practical Applications

Controlling Vocal Harshness and Sibilance

Vocal recordings often suffer from harsh resonances in the 2–5 kHz range and sibilance above 6 kHz. Instead of using a de‑esser (which is essentially a narrow‑band compressor), you can use a multi‑band compressor to target only the problematic frequencies. Set a band centred around 3–4 kHz with a fast attack, moderate ratio (3:1), and release that follows the natural decay of the voice. This reduces harshness without making the vocal sound thin or overly processed.

Taming the Low End Without Losing Punch

The low end (sub‑bass to 120 Hz) is often the most inconsistent part of a mix. A kick drum might have an overwhelming sub boom one moment and disappear the next. By compressing the low band separately (typically below 80–100 Hz) with a medium attack (10–15 ms) and a high ratio (4:1 to 6:1), you can even out low‑frequency peaks while preserving the transient attack of the kick. The mid and high bands remain unaffected, keeping the cymbals and vocals clear.

Balancing Dense Instrumentation

In a mix with many competing elements, multi‑band compression can act as a frequency‑dependent glue. For example, if your guitar and synth share the same midrange, you can compress that midrange band to reduce level conflicts. Similarly, if a piano part is too bright in the left channel, you can compress the high band on that side. This technique is often used in broadcast and podcasting to ensure consistent loudness across different frequency ranges.

Mastering and Loudness Maximisation

In mastering, multi‑band compression is used to control overall dynamic range while maintaining mix integrity. A very common approach is to apply a gentle 1.5:1 or 2:1 ratio on each band with slow attack and release to catch only the loudest peaks. This allows the limiter to work more transparently, increasing perceived loudness without obvious pumping. Many mastering engineers use a three‑band setup: low (20–150 Hz), mid (150 Hz–3 kHz), and high (3 kHz–20 kHz), with the low band often compressed slightly more to keep the bass consistent across playback systems.

Advanced Techniques and Workflow Tips

Side‑chain Input for Frequency‑Specific Ducking

Some multi‑band compressors allow you to side‑chain an external signal to a specific band. For example, you can route a kick drum to trigger compression only on the low band of the bass guitar. This creates a cleaner, tighter low end without the need for conventional side‑chain compression that ducks everything. This technique is popular in electronic music and modern pop production.

Using Linear‑Phase Crossovers for Mastering

As mentioned, linear‑phase crossovers avoid phase shift but introduce latency. For mastering (which is done offline), this is ideal because it preserves transient integrity. In mixing, if you need low latency for real‑time recording, use minimum‑phase crossovers and compensate with careful gain staging. Some plug‑ins offer a blend control that lets you dial in a mix of both crossover types.

Parallel Multi‑band Compression

Rather than applying compression 100% wet, you can mix the compressed signal in parallel with the dry signal on a per‑band basis. This allows you to keep the natural dynamics while adding the benefits of compression only where needed. Many plug‑ins have a wet/dry mix per band or a “mix” knob for the whole unit. Parallel multi‑band compression works exceptionally well for drum bus processing – compressing the low band heavily for punch and blending it back with the original punchy low end.

Gain Staging and Makeup Gain

Each band’s makeup gain must be set carefully. After compression, the level of each band may shift, causing tonal imbalances. Use an output meter to ensure that the overall level of each band equals the level of the original signal (when the compressor is bypassed). Some plug‑ins feature auto‑makeup gain, but it’s safer to adjust manually by ear. A common mistake is to over‑compress and then increase makeup gain, which introduces noise and distortion. Aim for 2–4 dB of gain reduction per band as a starting point.

Listening Critically and Using Visual Aids

Multi‑band compression can easily mask problems if you rely solely on meters. Always toggle the compressor on and off after each adjustment, and listen to the effect on the mix as a whole, not just soloed bands. Use the spectral display of your DAW to see how the frequency balance changes. Watch the gain‑reduction meters per band – if one band is compressing heavily while others are not, that band may need a different threshold or a more gentle ratio.

Common Pitfalls and How to Avoid Them

  • Over‑compression – Compressing each band too much leads to a lifeless, flat mix. Start with subtle ratios (1.5:1 to 2:1) and only increase if necessary.
  • Phase Issues at Crossover Points – Linear‑phase crossovers solve this but introduce latency. If you hear comb filtering or a “hole” in the frequency response, adjust crossover points or switch to a higher‑quality crossover algorithm.
  • Conflicting Bands – If your crossover points are too close, the compressors can fight each other. Leave a gap of at least one octave between crossovers unless you have a specific need for overlap.
  • Ignoring the Mix’s Natural Envelope – Multi‑band compression is not a substitute for good arrangement and mixing. If a track has extreme frequency imbalances, consider EQ or level automation first.

Choosing the Right Multi‑band Compressor

Many audio plug‑ins offer multi‑band compression, from stock DAW tools to premium third‑party options. Free options include the TDR Nova (a dynamic EQ with multi‑band capabilities), while popular paid choices include FabFilter Pro‑MB, iZotope Ozone Dynamics, and Waves C6. When selecting a plug‑in, consider factors like crossover quality, number of bands, side‑chain routing, and visual feedback. For beginners, starting with a three‑band setup is often enough.

If you are new to multi‑band compression, try applying it to a simple drum loop. Set the crossover at 100 Hz, compress the low band moderately, and leave the high band untouched. Then add a second crossover at 2 kHz, compress the mid band to control snare and hi‑hat overlap, and listen to how the low end stays tight while the high end remains open. This hands‑on experiment will quickly teach you how the tool behaves.

Conclusion

Multi‑band compression is a precise, powerful method for controlling dynamic range across different frequency regions. When used thoughtfully, it can clean up a muddy low end, tame harsh vocals, and bring clarity to dense arrangements without sacrificing natural dynamics. The key is to use it as a corrective tool rather than a cure‑all. Combine it with careful EQ, automation, and standard compression for the best results. Experiment with different crossover points, ratios, and attack/release settings, and always trust your ears over the meters.

As you become more familiar with multi‑band compression, you will find it invaluable for everything from tracking to final mastering. For additional reading, check out Sound On Sound’s guide to multi‑band compression and iZotope’s in‑depth tutorial. Practice on real mixes, and you will soon notice a significant improvement in your ability to control dynamics with precision and nuance.