How To Use A Spectrum Analyzer

Updated May 2026

Every experienced mixing engineer will tell you the same thing: you should always trust your ears first. But there's a reason every professional studio on the planet has at least one spectrum analyzer open during a session. These tools reveal things about your audio that ears alone — especially in imperfect listening environments — will miss. A rogue resonance at 400Hz, a build-up in the low-mid mud zone, a high-frequency shelf that's sucking the air out of a vocal: a spectrum analyzer catches all of this and makes it visible.

If you've been treating spectrum analyzers as decorative plugins that flash colorfully in the background, this guide is going to change the way you work. We'll cover the fundamental theory, the most important display settings, real-world workflows for mixing and mastering, and how to interpret what you're seeing to make smarter decisions at every stage of production.

What Is a Spectrum Analyzer and How Does It Work?

A spectrum analyzer is a metering plugin or hardware unit that displays the amplitude (level) of an audio signal across the frequency spectrum, typically from 20Hz to 20kHz — the range of human hearing. The horizontal axis represents frequency (usually on a logarithmic scale so that octaves appear equally spaced), and the vertical axis represents amplitude, typically in decibels (dBFS or dBr).

The underlying math is the Fast Fourier Transform, or FFT. When audio passes through a spectrum analyzer, the FFT algorithm decomposes the signal into its constituent frequencies and calculates the amplitude of each. The result is displayed as either a real-time curve that moves with the signal or as a static "frozen" snapshot you can examine at leisure.

FFT Size and Resolution

The FFT size — also called the block size or window size — is one of the most important settings you'll encounter. Common options range from 256 to 16,384 samples (and sometimes higher). Here's the trade-off:

• Small FFT size (256–1024): Fast temporal response — you see changes in the spectrum quickly. Low frequency resolution — frequency bins are wide, so close frequencies blend together. Useful for watching transient-heavy material like drums.
• Large FFT size (4096–16384): Slow temporal response — the display lags behind the audio slightly. High frequency resolution — individual frequencies are clearly separated. Ideal for identifying narrow resonances and harmonic content.

For most mixing work, an FFT size of 2048 to 4096 is a good default. When hunting for specific problem resonances — a ringing in a vocal, a tuned room mode in a kick — bump up to 8192 or 16384 for the extra resolution.

Window Functions

The FFT window function determines how the analyzer handles the edges of each analysis block. Options you'll typically see include Hann (also spelled Hanning), Blackman-Harris, Flat Top, and rectangular. For general audio analysis, Hann is the standard choice: it offers a good balance between frequency resolution and spectral leakage. Flat Top windows give the most accurate amplitude readings, which is why they're preferred for calibration work. Blackman-Harris offers even better leakage suppression at the cost of slightly wider main lobes. For 99% of music production tasks, leave the window function on Hann and focus your attention elsewhere.

Averaging Modes

Raw FFT data is extremely jittery — the spectrum bounces constantly, making it hard to read. Averaging smooths this out by blending multiple analysis frames together. You'll typically see two types:

• Temporal averaging: Averages over time, producing a more stable curve. Higher averaging values give a smoother display but respond more slowly to changes.
• Spectral averaging (RMS mode): Averages the energy across neighboring frequency bins, producing a smoother curve across the frequency axis.

Many analyzers also have a "peak hold" function that freezes the highest level each frequency bin has reached. This is extremely useful during mastering — run your entire mix through and the peak hold curve shows you the overall spectral shape of the song without you having to watch it in real time.

Key Display Settings Every Producer Should Understand

Modern spectrum analyzer plugins — tools like FabFilter Pro-Q 3, iZotope Insight 2, Voxengo SPAN, and Waves PAZ Analyzer — offer a wide range of display configurations. Knowing what each setting does prevents you from misreading the data you're looking at.

Display Scale: Linear vs. Logarithmic Frequency Axis

The frequency axis of a spectrum analyzer should almost always be displayed on a logarithmic scale for music production work. Here's why: human pitch perception is logarithmic. The octave from 100Hz to 200Hz is perceptually the same "width" as the octave from 4kHz to 8kHz — but if you displayed this on a linear scale, the bass octave would be crammed into a tiny sliver while the high frequencies would stretch across most of the screen. On a log scale, each octave gets roughly equal horizontal real estate, which matches how we hear.

Some analyzers allow you to switch to a linear frequency scale. This is occasionally useful for specific calibration or scientific measurement tasks, but for mixing decisions, always use logarithmic.

Amplitude Scale and Range

The vertical axis shows amplitude in dBFS (decibels relative to full scale). The range you display matters. A narrow range (say, -40dBFS to 0dBFS) gives you more visual detail in the loud parts of the spectrum but clips off low-level content. A wider range (-120dBFS to 0dBFS) shows everything but compresses the visual resolution.

For mixing: a range of -90dBFS to 0dBFS is practical — wide enough to catch noise floor issues, tight enough to be readable. For mastering and loudness analysis, you might narrow this to -60 to 0dBFS to better see the shape of the program material.

Slope Compensation (Pink Noise Reference)

This is one of the most misunderstood settings in spectrum analysis. A well-balanced mix, when analyzed against a flat frequency axis, naturally slopes downward from low to high frequencies. This is because pink noise — the acoustical model for balanced audio content — has 3dB more energy per octave as you go lower in frequency.

Many analyzers include a "pink noise slope" or "tilt" setting (often labeled as +3dB/octave or -3dB/octave depending on the direction) that compensates for this. When this compensation is applied, a perfectly balanced pink noise signal appears as a flat horizontal line instead of a downward-sloping one. This makes it much easier to see whether your mix has too much or too little energy in a given band relative to a spectrally balanced reference.

Plugins like Voxengo SPAN display this as a default slope. The FabFilter Pro-Q 4 spectrum display also allows you to adjust the tilt of the reference line. Understanding this prevents you from looking at a naturally sloping mix curve and incorrectly diagnosing a low-end problem.

Channel Configuration: Stereo, Mid, Side

Advanced analyzers can display the spectrum of the mid channel (sum of left and right, representing the mono content) separately from the side channel (the difference between left and right, representing stereo width). This is invaluable for checking that your low frequencies are mono — energy below roughly 80–100Hz in the side channel is problematic for vinyl cutting and can cause phasing issues on mono playback systems. It's also useful for identifying whether reverb tails and ambience (which should live primarily in the side channel) are at appropriate levels relative to your direct, centered content.

How to Place Spectrum Analyzers in Your Session

Knowing where to insert a spectrum analyzer is as important as knowing how to read it. Different placements reveal different types of information.

On Individual Tracks

Placing a spectrum analyzer on an individual track — say, a bass guitar, kick drum, or lead vocal — shows you the raw frequency content of that instrument in isolation. This is useful for:

• Identifying resonances: A narrow spike in the 300–500Hz range often indicates a boxiness resonance in a vocal booth or a body resonance in a guitar. You'll see it as a sharp peak standing above the surrounding spectrum.
• Finding the fundamental frequency: The fundamental note of a bass guitar or kick drum appears as the first (and usually largest) peak in the low end. Knowing exactly where it sits — 50Hz vs 80Hz, for instance — tells you precisely where to focus your bass EQ.
• Checking harmonic relationships: Harmonics appear at integer multiples of the fundamental. A kick with a fundamental at 60Hz will have harmonics at 120Hz, 180Hz, 240Hz, and so on. This helps you understand why certain frequencies feel connected or conflicting across instruments.

On Subgroups and Buses

Insert an analyzer on your drum bus, instrument bus, or vocal bus to see how a group of related elements combines. This helps you catch issues that only appear in the summed signal — like a frequency buildup at 200Hz when the kick, bass, and low-mid heavy guitars are all contributing energy in the same band.

On the Master Bus

This is the most common placement for spectrum analyzers. With an analyzer on the master bus, you can see the overall spectral balance of your entire mix in real time. Compare the shape of your mix curve to reference tracks (professional releases in the same genre) to identify where your mix diverges from what's expected. This is especially helpful if you're working in a room with significant acoustic problems — the analyzer gives you objective data that your room's coloration might be masking.

Using Two Analyzers for A/B Comparison

One of the most powerful workflows is to run two spectrum analyzers simultaneously — one on your mix and one on a reference track — and switch between them (or use an analyzer that allows you to load a reference trace). FabFilter Pro-Q 4 lets you freeze a spectrum snapshot and display it alongside the live signal, while iZotope Insight 2 allows multi-source metering. This tells you at a glance whether your kick drum has as much sub energy as the reference, whether your vocal sits at a comparable level in the 2–5kHz presence region, and whether your master has a similar high-shelf rolloff.

Pre vs. Post EQ Placement

If you're using an EQ with a built-in spectrum display — like the FabFilter Pro-Q 3 or Pro-Q 4 — you have the option to view the spectrum pre- or post-EQ processing. Viewing the pre-EQ spectrum tells you what the signal looks like before your processing, which is what you're responding to. The post-EQ view shows you what you're sending downstream. Many engineers keep the pre-EQ spectrum visible while drawing EQ curves so they can see exactly what they're cutting or boosting relative to the original signal.

Reading the Spectrum in Context: What You're Actually Looking At

The most common mistake beginners make is treating the spectrum analyzer as something to "fix" — trying to make the curve look like a straight line or match some imagined ideal shape. That's not how it works. The spectrum display is a diagnostic tool, and reading it correctly means understanding what's normal, what's problematic, and what's a stylistic choice.

Understanding Frequency Zones

Before you can read a spectrum, you need to internalize the frequency zones and what they contain musically:

• Sub-bass (20–60Hz): The deep rumble and fundamental energy of kick drums and bass instruments. Felt more than heard on most systems. Too much here causes a muddy, unfocused low end; too little leaves the mix sounding thin on subwoofers and club systems.
• Bass (60–250Hz): Body of bass guitar, lower harmonics of kick, warmth of piano and acoustic guitar. This region determines whether a mix feels fat and full or lean and tight. A build-up at 100–200Hz is the classic source of "mud."
• Low-mids (250Hz–500Hz): The boxy, honky zone. Most problematic resonances live here. Male vocal chest resonance, guitar body boom, room reflections — if your mix sounds congested or claustrophobic, this is usually why.
• Mids (500Hz–2kHz): Presence and intelligibility of vocals, attack of guitar, definition in piano. This is where the human ear is most sensitive. Too much adds nasal harshness; too little makes the mix sound distant and hollow.
• Upper-mids (2kHz–5kHz): The "presence" zone. This is where vocals cut through a mix, where guitars gain bite, and where harshness or fatigue accumulates with prolonged exposure. Boosting here adds aggressiveness; cutting adds smoothness.
• Presence/brilliance (5kHz–10kHz): Sibilance lives at 6–8kHz. Cymbals, string attack, vocal brightness. Critical for perceived clarity and definition on consumer playback systems.
• Air (10kHz–20kHz): Shimmer, sparkle, the sense of space. Subtle presence here adds professional sheen; too much creates harshness; too little makes a mix sound dull and old.

Normal Spectral Shapes for Common Genres

Different genres have characteristic spectral shapes. When you compare your mix against a professional reference, you're checking alignment with these norms — not trying to copy them exactly, but understanding where intentional differences lie versus accidental imbalances.

Hip-hop and trap mixes typically show a significant sub-bass presence (40–80Hz), a relatively flat mid-range, and a moderate-to-bright top end. EDM and electronic dance music often exhibits heavy sub content, a pronounced mid-range dip in the 300–500Hz range (to keep the mix clean and punchy), and extended, boosted highs. Rock mixes typically show a mid-forward character with a pronounced 2–5kHz range for guitar aggression and vocal cut. Orchestral and classical recordings tend to have a gradual rolloff above 8kHz and more balanced energy across the low-mids. Understanding these tendencies helps you read your analyzer in context rather than in the abstract.

Identifying Common Problems Visually

Here are the most common spectral problems and what they look like on an analyzer:

• Low-end mud: A broad hump of excess energy centered around 150–300Hz. The curve rises well above your reference in this zone and doesn't fall away quickly. Fix: broad EQ cut in this area, or high-pass filtering individual tracks contributing to the buildup.
• Boxiness: A narrower peak, often sharper than mud, typically in the 300–600Hz range. Common on recorded acoustic instruments, room-heavy vocal performances, and budget condenser microphones. Fix: identify the exact frequency with a narrow resonant boost as a "search" move, then apply a narrow cut once you find it.
• Harshness: A broad rise in the 2–5kHz range that doesn't come down. The curve stays elevated through the presence zone. Fix: gentle broad EQ cut, or dynamic EQ/multiband compression in this range.
• Sibilance spike: A narrow peak at 6–8kHz on a vocal or high-hat track. Extremely common on condenser microphone recordings. Fix: de-esser or narrow dynamic EQ cut triggered at that frequency.
• Thin low end: The curve drops off sharply below 100Hz, significantly below your reference. The mix will sound brittle on full-range systems. Fix: check high-pass filter settings across your individual tracks — many producers high-pass too aggressively.
• Dull/dark mix: The spectrum rolls off significantly above 8–10kHz compared to your reference. The mix will lack air and definition. Fix: gentle high-shelf boost on the master or individual tracks, or check whether overuse of analog-modeled saturation is causing top-end rolloff.

Using the Analyzer During EQ Work

When applying EQ, use the spectrum analyzer as follows: start your EQ search by ear — identify something that sounds problematic, then look at the analyzer to confirm which frequency region is involved. Apply your EQ cut or boost while watching the analyzer to confirm that the change is affecting the correct region and that you're not introducing new problems in adjacent bands. This ear-then-eyes workflow prevents you from chasing visual "perfection" at the expense of what actually sounds good.

This approach is central to good vocal mixing technique — you identify the harshness by ear first, then use the analyzer to pinpoint whether it's at 3kHz, 4kHz, or 5kHz before committing to a cut.

Using Spectrum Analyzers for Mastering and Delivery

Spectrum analyzers become even more critical at the mastering stage, where small spectral imbalances that survive mixing can be caught and corrected before a track is delivered to streaming platforms, distributors, or vinyl pressing plants.

Reference Track Comparison

The reference track workflow is one of the most valuable uses of a spectrum analyzer in mastering. Bring in a professionally mastered track in the same genre, match the loudness between it and your master (using an integrated loudness meter — LUFS, not peak level), and then compare the spectral shapes side by side. Look for:

• How does the low-end slope compare? Is your master heavier or lighter below 80Hz?
• Is there a dip in the 3–5kHz range in the reference (often intentional in certain pop and R&B masters) that your mix doesn't have?
• Does the reference extend more brightly above 12kHz? This is common in well-mastered pop and electronic music.

The goal is not to make your master identical to the reference, but to understand the intentional differences versus unintentional imbalances.

Streaming Platform Loudness Compliance

When mastering for streaming platforms like Spotify, Apple Music, and Tidal, loudness normalization (typically to -14 LUFS integrated on Spotify, -16 LUFS for podcasts on Apple) means that excessively loud masters will be turned down. A spectrum analyzer used in conjunction with a loudness meter helps you check that the spectral balance doesn't shift dramatically as you push the limiter harder. Aggressive limiting tends to compress the dynamic range unevenly, often dulling transients in the 3–8kHz range and adding low-mid buildup. Watch the spectrum before and after your limiter to catch these artifacts.

Mid/Side Analysis for Stereo Integrity

Switch your analyzer to mid/side mode and examine the side channel spectrum carefully below 100Hz. Any significant energy in the side channel below this frequency indicates a stereo low end — bass content that exists differently in the left and right channels. This causes phase cancellation on mono playback and is a problem for streaming, broadcast, and vinyl. Look for the side channel spectrum to be effectively silent below 80–100Hz on a well-mastered record. If it isn't, apply a mid/side EQ or low-pass filter on the side channel to mono the low end.

High-Frequency Air Shelf and the Nyquist Limit

On the high end, watch for a sharp rolloff at or near 20kHz. This can indicate that a track was processed or recorded at a sample rate that's limiting the high-frequency ceiling, or that a previous generation of bounce had a low-pass filter applied. Some mastering engineers apply a subtle high-shelf boost above 16kHz — the "air band" — to restore a sense of openness. The spectrum analyzer confirms whether this shelf is actually adding energy (the curve rises) or whether you've hit a hard bandwidth limit where boosting has no effect.

Using Spectrum Analysis in Atmos and Spatial Mixes

For producers working on Dolby Atmos and spatial audio deliveries, spectrum analysis becomes a multi-channel task. Individual bed channels and object tracks need to be analyzed both in isolation and in the rendered binaural output. Pay particular attention to the binaural render's spectral consistency across the soundstage — HRTF processing can create unpredictable frequency colorations, particularly in the 2–8kHz range where pinnae-related cues operate. Checking the spectrum of the binaural bounce against your stereo master is an important QC step in any Dolby Atmos mixing workflow.

The Best Spectrum Analyzer Plugins and What Makes Them Different

The market for spectrum analyzers ranges from free, lightweight utilities to comprehensive metering suites with loudness, phase, and correlation analysis built in. Here are the most important options as of May 2026:

FabFilter Pro-Q 4

The FabFilter Pro-Q 4 is primarily an EQ, but its built-in spectrum analyzer is among the most useful in the industry. It features a high-resolution real-time spectrum display, a spectrum grab function for comparing pre- and post-EQ, a collision detection system that highlights frequency masking between tracks when you have multiple Pro-Q instances open, and the ability to display the spectrum of other tracks for mid/side analysis. The Pro-Q 4's spectrum is tightly integrated with the EQ workflow — hovering over a peak in the spectrum display automatically suggests a matching EQ band — which makes it an indispensable tool for mixing engineers who are used to working visually as well as aurally. $179

Voxengo SPAN (Free)

Voxengo SPAN is the industry-standard free spectrum analyzer, available in both 32-bit and 64-bit formats for Mac and PC. It offers highly configurable FFT size, averaging, and slope settings, supports stereo and mid/side analysis, and displays up to eight channels simultaneously with user-defined colors. For producers on a budget, SPAN provides professional-grade analysis at no cost. Its display style is more utilitarian than FabFilter's polished interface, but the underlying analysis engine is extremely accurate and trusted in professional contexts.

iZotope Insight 2

iZotope Insight 2 is a comprehensive metering suite that includes a spectrum analyzer, loudness meter (LUFS, RMS, peak), spectrogram, stereo field visualizer, and phase correlation meter in a single plugin. The spectrum analyzer component supports up to six sources simultaneously, making it ideal for detailed reference track comparison. Insight 2 is particularly popular in post-production and broadcast contexts where multiple loudness standards (EBU R128, ATSC A/85, ITU-R BS.1770) need to be monitored alongside spectral balance. $99 (often available in iZotope bundles)

Waves PAZ Analyzer

The Waves PAZ Analyzer combines a spectrum display with a stereo position meter and peak/RMS loudness readouts in a single window. It's not as visually sophisticated as Pro-Q or Insight 2, but it remains widely used because of its simplicity and the fact that it's included in nearly every Waves bundle. If you already have Waves plugins, PAZ is a capable, zero-cost addition to your metering setup. $29 (often included in bundles)

Melda Production MAnalyzer (Free)

Melda's MAnalyzer is a highly capable free option with an unusually deep feature set for a free plugin. It supports A/B spectrum comparison, a spectrogram display, sonogram mode, and extensive customization of display parameters. The interface is complex but rewards the investment of time spent learning it. For producers who need more analytical depth than SPAN offers but don't want to spend money on iZotope Insight, MAnalyzer is a strong choice.

The Spectrogram vs. the Spectrum Display

A distinction worth making: the spectrum analyzer displays amplitude vs. frequency at a given moment in time. A spectrogram displays frequency on the vertical axis and time on the horizontal axis, with amplitude encoded as color intensity. Spectrograms are excellent for viewing the evolution of spectral content over the course of a recording — identifying where a rumble enters, how long a room resonance sustains, whether a vocal has consistent brightness throughout a performance, or where noise artifacts occur. Many of the plugins above include both display types. Learn to read both — they provide complementary perspectives on the same information.

Practical Workflow Examples: Spectrum Analyzers in Action

Theory only gets you so far. Here's how professional-level spectrum analyzer usage looks in actual production workflows.

Workflow 1: Fixing a Muddy Mix

You've built up a full arrangement — drums, bass, guitars, synths, vocals — and the mix sounds congested and heavy. Step one: insert a spectrum analyzer (Voxengo SPAN or Pro-Q 4 spectrum view) on the master bus and look at the 150–400Hz region. If there's a broad hump here rising 6–10dB above your pink-noise-compensated reference line, you have a classic low-mid buildup problem.

Now go track by track. Solo each instrument and look at what it contributes to this region. Bass guitar, kick drum, electric guitar, piano, and synth pads can all simultaneously contribute energy at 200Hz without any single one of them being "wrong" in isolation. The problem only appears in the sum. Apply subtle high-pass filtering on each non-bass element to roll off their low-mid contributions. A high-pass at 80–120Hz on guitars, at 150–200Hz on synth pads, at 100Hz on piano usually clears significant mud without audibly thinning any individual instrument. After making these cuts, re-examine the master bus spectrum — the hump should shrink toward the reference line.

Workflow 2: Matching a Reference Track

You're finishing a track and want to verify that it translates to professional playback standards. Bring a reference track into your DAW on a separate audio track, level-match it to your mix using an integrated LUFS meter, and insert SPAN (or your analyzer of choice) set to receive input from both tracks via two separate instances or a multi-source analyzer. Enable peak hold and play the reference track for its full duration, then do the same with your mix.

Compare the resulting curves. Common findings: your mix has 4–6dB more energy at 100–200Hz (too much warmth/mud), the reference has 3–5dB more energy at 8–12kHz (your mix is too dark), or the reference has a distinctive dip at 3kHz that your mix doesn't (your mix is more aggressive/harsh). Each of these findings translates directly into EQ decisions on your master bus.

Workflow 3: Identifying Resonances in a Recorded Vocal

You've recorded a lead vocal that sounds great in isolation but unpleasant and honky in the mix. Insert a high-resolution analyzer on the vocal track with a large FFT size (8192 or 16384) and enable a long averaging time. Play a sustained vowel section of the vocal and look for narrow peaks standing up above the surrounding spectrum. Resonances typically appear as sharp, narrow spikes rather than broad humps. Common locations: 200–300Hz (chest resonance, microphone proximity effect), 400–600Hz (boxy room reflections), 800Hz–1.2kHz (nasal resonance), 3–4kHz (harshness resonance from condenser microphone response curves).

Once you've identified the resonance visually, use a narrow EQ band (Q of 4–8) to apply a 4–8dB cut at that exact frequency. Listen carefully — you should hear the problematic character reduce without the vocal losing body or presence. For moving vocal resonances that shift in pitch over the course of a performance, consider a dynamic EQ that only engages the cut when the resonance exceeds a certain threshold.

Workflow 4: Checking Low-End Mono Compatibility

Switch your spectrum analyzer to mid/side mode and look at the side channel spectrum below 100Hz. Any sustained energy in this region represents stereo content in your low end. On a professional dance music master, the side channel should be essentially empty below 80Hz. If it isn't, identify the source: stereo synthesizer bass patches, wide stereo enhancement plugins processing the whole mix, or stereo room microphones blending into the drum overheads are the most common culprits. Solve the problem by either applying a narrow low-pass filter to the side channel on your master bus (cutting everything below 80–100Hz in side only), or by addressing stereo imaging at the source tracks.

Workflow 5: Tracking Down a High-Frequency Harshness

A client's mix sounds harsh and fatiguing at a certain point in the arrangement when the full band is playing. Insert the analyzer on the master bus and play the problematic section. Note the frequency range that rises during that section — often you'll find a 2–5kHz rise that correlates with a particular instrument entering the arrangement (a distorted guitar, a synth lead, a bright snare top microphone). Solo or mute elements one by one while watching the spectrum to identify which track is contributing the harshness. Once identified, apply a targeted cut on that track's EQ rather than cutting the entire master — this preserves detail in all the other elements that don't have the same issue.

This kind of precise diagnostic workflow is exactly what separates producers who mix with intention from those who apply broad, imprecise corrections that fix one problem while creating others. Developing this skill — using the analyzer as a precision diagnostic tool rather than a visual guide to chase — is the foundation of a professional metering practice.

Workflow 6: Checking Beat Frequency Content Before Sending to Vocalists

Before sending an instrumental beat for a vocalist to record over, run the spectrum analyzer on the mix and check the 2–5kHz range. If the beat has significant energy in this zone — distorted hi-hats, aggressive synth leads, overdriven 808s with lots of upper harmonics — the vocalist will have to fight for space in the presence zone. This creates headaches in mixing later. Identify these competing elements early, apply a gentle roll-off in the upper-mids on the beat, and leave space for the vocal to occupy the 2–5kHz range by design rather than by post-mix surgery. This principle applies across all genres and is especially important in trap and hip-hop production where 808s can carry significant harmonic energy through the presence zone.

Practical Exercises

Frequently Asked Questions