Lesson Objective

This lesson explains multi-band processing — the technique of splitting an audio signal into separate frequency bands and applying independent dynamics processing to each band. You will learn how multi-band compressors and expanders work, how to set crossover frequencies intelligently, and when multi-band processing is the right choice versus simpler single-band alternatives. By the end, you will be able to use multi-band tools confidently in both mixing and mastering contexts.

What You Will Learn

  • How multi-band processing splits audio into frequency bands for independent treatment
  • The signal flow inside a multi-band compressor
  • How to set crossover frequencies based on the audio content
  • Multi-band expansion and its applications
  • When to choose multi-band over single-band compression
  • Common multi-band applications in mixing: taming problem frequencies dynamically
  • Multi-band compression in mastering: balancing the spectral dynamics of a full mix

Required Knowledge or Tools

Multi-band processing is an advanced topic that requires a thorough understanding of both compression and equalization. You should be comfortable with all standard compressor parameters and understand how frequency ranges relate to musical content before attempting multi-band work.

  • Completion of Lesson 7 (Equalization) and Lesson 8 (Compression)
  • A multi-band compressor plugin (FabFilter Pro-MB, iZotope Ozone, Waves C4, or DAW stock options)
  • A multi-track session or stereo mix to practice on
  • Spectrum analyzer for visual feedback on frequency content

Core Concept Explanation

A standard compressor treats the entire audio signal as a single entity. When a loud bass note triggers the compressor, the gain reduction affects the entire frequency spectrum — the highs get quieter along with the lows. This can cause the mix to feel like it is pumping or losing clarity whenever the low end is active. Multi-band processing solves this by dividing the signal into separate frequency bands and compressing each band independently.

Signal Flow in a Multi-band Compressor

The signal enters the multi-band processor and is split by crossover filters into separate frequency bands — typically three to five bands, though some processors offer more. Each band passes through its own independent compressor with its own threshold, ratio, attack, release, and makeup gain settings. After processing, the bands are recombined and summed back into a single signal at the output.

The crossover filters are the critical element. They determine which frequencies belong to which band. Typical band divisions for a four-band processor might be: low band (20–200 Hz), low-mid band (200–2,000 Hz), high-mid band (2,000–8,000 Hz), and high band (8,000–20,000 Hz). These divisions are adjustable and should be set based on the specific content being processed.

Key Insight: Multi-band compression is essentially dynamic EQ. When a specific frequency range gets too loud, the compressor for that band reduces its level. When it quiets down, the band returns to its normal level. This is fundamentally different from static EQ, which applies the same cut or boost regardless of the signal level.

Setting Crossover Frequencies

Crossover frequency placement is one of the most important decisions in multi-band processing. Poorly placed crossovers can cause phase issues, unnatural tonal shifts, and processing artifacts. The goal is to place crossovers in frequency regions where the audio content is relatively sparse, so that the transition between bands is smooth and the processing of one band does not audibly affect adjacent bands.

Common crossover placements for music mastering are around 80–120 Hz (separating sub bass from bass), 300–500 Hz (separating bass from midrange), and 3,000–5,000 Hz (separating midrange from high frequencies). These are starting points — always adjust based on the specific material you are working with.

Multi-band Expansion

Multi-band expanders work on the same principle as multi-band compressors but apply expansion rather than compression to each band. When a frequency band drops below the threshold, the expander reduces its level further, effectively increasing the dynamic range of that band. This is useful for reducing low-level noise or rumble in specific frequency ranges without affecting the rest of the signal, and for adding punch and transient clarity to specific frequency bands.

Some multi-band processors allow you to switch individual bands between compression and expansion mode, giving you a highly flexible tool that can simultaneously compress one frequency range and expand another.

Multi-band vs. Single-band: Choosing the Right Tool

Single-band compression is simpler, more transparent, and generally preferable when the dynamics problem affects the entire frequency spectrum equally. Use single-band compression for controlling the overall dynamic range of a track, adding punch and character, or gluing elements together.

Multi-band compression is the better choice when the dynamics problem is frequency-specific. A bass guitar that has inconsistent low-end energy but a consistent midrange benefits from multi-band compression on the low band only. A mix that sounds balanced most of the time but gets muddy when the bass and kick play together benefits from multi-band compression targeting the low-mid range.

Less Is More: Multi-band compression is easy to overuse. Start with minimal settings — low ratios, high thresholds, and gentle gain reduction of 2–4 dB maximum per band. Aggressive multi-band compression can make audio sound unnatural, lifeless, and over-processed. Use it as a surgical tool, not a blunt instrument.

Multi-band Processing in Mastering

In mastering, multi-band compression is used to balance the spectral dynamics of a finished mix. If the low end of a mix is inconsistent — sometimes too heavy, sometimes too thin — a multi-band compressor on the low band can smooth it out without affecting the midrange or high frequencies. Similarly, if the high frequencies are harsh on certain notes or instruments, a multi-band compressor targeting that range can tame the harshness dynamically.

Modern mastering tools like iZotope Ozone combine multi-band compression with dynamic EQ, spectral analysis, and AI-assisted processing to give mastering engineers unprecedented control over the spectral balance of a mix.

Visual Explanation

Multi-band compressor plugin interface

A multi-band compressor interface shows separate compression controls for each frequency band, with crossover points that can be adjusted to define the boundaries between bands.

Most multi-band compressor interfaces display a frequency spectrum view with vertical lines marking the crossover points between bands. Each band has its own set of compression controls, and gain reduction meters show how much compression is being applied to each band in real time. This visual feedback is essential for understanding what the processor is doing and making informed adjustments.

Why This Lesson Matters

Multi-band processing is a cornerstone of professional mastering and an increasingly important tool in mixing. As music production has moved to digital environments where the full frequency spectrum is always available, the need to manage frequency-specific dynamics has grown. A mix that sounds balanced on studio monitors may have an inconsistent low end that only becomes apparent on different playback systems. Multi-band compression can address these issues in ways that static EQ cannot.

Understanding multi-band processing also deepens your understanding of the relationship between frequency content and dynamics — two of the most fundamental aspects of audio. When you can think about these dimensions simultaneously and apply processing that addresses both at once, your ability to solve mixing and mastering problems improves significantly.

Common Pitfall: Multi-band compression can introduce phase artifacts at the crossover points, especially with linear-phase crossover filters. Always listen carefully to the crossover regions and compare the processed signal to the unprocessed original to ensure the processing is improving rather than degrading the audio.

Step-by-Step Tutorial

Follow this workflow to apply multi-band compression to a stereo mix bus:

  1. Analyze the Mix First: Before inserting any processing, use a spectrum analyzer to study the frequency content of your mix. Identify any frequency ranges that seem inconsistent — areas that are sometimes too prominent and sometimes too thin. This analysis guides your crossover placement and band settings.
  2. Insert the Multi-band Compressor: Place the multi-band compressor on your mix bus or master channel. Start with the plugin's default settings or a gentle preset as a starting point. Make sure the output gain is set to unity so you can compare the processed and unprocessed signal at the same level.
  3. Set Crossover Frequencies: Adjust the crossover points to match the content of your mix. For most music, start with crossovers around 100 Hz, 500 Hz, and 4,000 Hz. Listen to the mix and adjust the crossovers so they fall in frequency regions where the content is relatively sparse.
  4. Set Thresholds for Each Band: For each band, lower the threshold until the compressor just begins to engage on the loudest moments in that frequency range. You should see 2–4 dB of gain reduction at most. If you are seeing more than 6 dB of gain reduction, the threshold is too low.
  5. Adjust Attack and Release Per Band: Low bands generally benefit from slower attack times (20–50 ms) to preserve the punch of kick drums and bass transients. High bands can use faster attack times (5–15 ms) to control harshness more quickly. Set release times so the compression recovers naturally without pumping.
  6. Compare and Refine: Bypass the multi-band compressor and compare the processed and unprocessed signal. The processed version should sound more balanced and controlled without sounding smaller or less dynamic. If the processing is audible in a negative way, reduce the amount of gain reduction in the offending band.

Common Mistakes and Misunderstandings

Mistake 1: Using multi-band compression to fix problems that should be fixed at the source. If a specific instrument has an inconsistent frequency response, address it with EQ or compression on that track before reaching for multi-band processing on the mix bus.

Mistake 2: Setting crossovers at fixed, arbitrary frequencies without considering the content. Crossovers placed in the middle of an instrument's fundamental frequency range can cause that instrument to sound unnatural as different parts of its spectrum are processed independently.

Mistake 3: Applying too much gain reduction. Multi-band compression with more than 6 dB of gain reduction per band will almost always sound unnatural and over-processed. Keep gain reduction gentle and use the tool for control, not dramatic level changes.

Mistake 4: Ignoring the makeup gain. Each band has its own makeup gain control. If you are applying significant compression to one band, you may need to add makeup gain to that band to maintain the overall spectral balance of the mix.

Mistake 5: Confusing multi-band compression with dynamic EQ. While they are conceptually similar, they work differently. Dynamic EQ applies a boost or cut that scales with the signal level. Multi-band compression applies gain reduction above a threshold. Understanding the difference helps you choose the right tool for each situation.

Practical Example or Scenario

A mastering engineer receives a mix of an R&B track. The mix sounds great overall, but the low end is inconsistent — the bass guitar and kick drum create a heavy, boomy low end during the chorus, while the verses feel thin and lacking in weight. Static EQ cannot solve this because the problem is dynamic, not tonal.

She inserts a multi-band compressor on the master channel and sets a crossover at 120 Hz to isolate the low band. She sets the low band compressor to a ratio of 3:1 with a threshold that engages during the heavy chorus sections, applying about 3 dB of gain reduction. The attack is set to 30 ms to preserve the kick drum's punch, and the release is set to 150 ms for a natural recovery.

The result is a more consistent low end throughout the track. The chorus still feels powerful, but the excessive boominess is controlled. The verses feel fuller because the compressor is not engaging during the lighter sections, allowing the natural low-end content to come through unprocessed.

She also adds a gentle high-band compressor targeting frequencies above 5,000 Hz to tame occasional harshness in the cymbals and hi-hats. With 2 dB of gain reduction at most, the high end becomes smoother and more consistent without losing its air and presence.

Lesson Summary

Multi-band processing divides audio into separate frequency bands and applies independent dynamics processing to each. This allows you to address frequency-specific dynamics problems that single-band compression cannot solve without affecting the entire frequency spectrum.

Crossover frequency placement is critical and should be based on the content being processed. Keep gain reduction gentle — 2–4 dB per band is usually sufficient. Multi-band compression is most valuable when the dynamics problem is frequency-specific, while single-band compression remains the better choice for overall dynamic control.

The next lesson explores Harmonic Saturation and Distortion, covering how adding controlled harmonic content can bring warmth, character, and presence to digital audio productions.