what is resonance

the frequency that will not stop ringing

clap your hands in an empty room. you hear the clap, then you hear the room: a short burst of ringing that decays over a fraction of a second. that ringing is resonance. certain frequencies in the room are amplified by the walls, floor, and ceiling, and they sustain longer than everything else.

resonance in audio is exactly this: specific frequencies that build up louder or ring longer than the surrounding content. every physical system has resonant frequencies. your recording room, your microphone, your guitar body, your vocalist’s chest cavity. the question is never whether resonance exists. it is whether the resonances in your signal are helping or hurting.

when they help, resonance is what makes an acoustic guitar sound full and a vocal sound rich. when they hurt, resonance is the boomy low-mid build-up that makes a room recording sound boxy, or the harsh 3 kHz peak that makes a vocal fatiguing to listen to for more than 30 seconds.

why things resonate

resonance happens when a system’s physical properties cause it to amplify certain frequencies.

take a room. sound bounces between parallel walls. at frequencies where the distance between the walls is a whole number of half-wavelengths, the reflected waves reinforce each other. a room that is 3.4 meters wide has a fundamental resonant mode at approximately 50 Hz (the wavelength of 50 Hz is about 6.8 meters, and half of that fits exactly between the walls). it also resonates at 100 Hz, 150 Hz, 200 Hz, and so on, at integer multiples of the fundamental.

these are room modes, and they are the most common source of low-frequency resonance problems in home studios. a bass note that hits one of these modes sounds louder and rings longer than neighboring notes. the room is adding energy at that frequency.

instruments work the same way. a guitar body is a resonant cavity. the size and shape of the body determine which frequencies it amplifies. this is what gives different guitar shapes their different tonal characters. a dreadnought resonates differently from a parlor guitar, which resonates differently from a classical.

what resonance looks like in your mix

in a spectrum analyzer, resonance shows up as narrow peaks that persist regardless of what is playing. they stand out from the surrounding content because they ring at a constant frequency while the music moves around them.

there are three frequency ranges where resonance causes the most trouble in a typical mix:

80-250 Hz: boxiness and boom

this is where room modes live. if you recorded in an untreated room, you almost certainly have resonant build-up somewhere in this range. it sounds boxy, muddy, or boomy. one bass note sounds much louder than its neighbors. a kick drum rings at a frequency the room is amplifying.

250-800 Hz: nasal and honky

the “cardboard box” frequencies. resonance here makes things sound cheap, thin, and hollow. vocal recordings in small rooms often have a nasal quality from room modes in this range. acoustic guitars recorded close to a wall can pick up resonant coloring here.

2-5 kHz: harsh and fatiguing

the most immediately painful range. your ears are more sensitive between 2 and 5 kHz than anywhere else in the spectrum.[^1] a 3 dB resonant peak at 3 kHz is far more noticeable than the same peak at 300 Hz. vocal harshness, guitar bite, cymbal brittleness: most “it hurts to listen to this” problems live here.

resonance vs character

here is where it gets nuanced. not all resonance is bad.

the body of an acoustic guitar IS a resonant system. remove the resonance and you remove the guitar’s tone. a singer’s voice IS a product of resonant cavities in the throat, mouth, and nasal passages. a drum shell rings at its fundamental. these resonances are not problems. they are the instrument.

the skill is distinguishing between resonance that serves the music and resonance that distracts from it.

a useful test: does this frequency sound intentional? a guitar body resonance at 200 Hz that gives the instrument warmth and projection is musical. a room mode at 120 Hz that makes the low end boom on every other note is not. a vocal presence at 3 kHz that helps the singer cut through the mix is intentional. a harsh peak at 3.5 kHz from a cheap microphone capsule is not.

where resonance comes from in practice

in a typical production workflow, resonance enters your signal at multiple points:

the room

every untreated room has resonant modes. bass traps and acoustic treatment reduce these, but most home studios still have significant resonances below 300 Hz. the smaller the room, the higher the fundamental mode frequency, and the more audible the problem.

the microphone

condenser microphones are designed with a presence peak between 3-5 kHz. this is intentional: it adds clarity and “air” to recordings. but when combined with a source that already has energy in that range (a bright vocalist, a close-miked acoustic guitar), the mic’s resonance stacks on top of the source’s resonance.

the signal chain

preamps, compressors, and EQ all have frequency-dependent behavior. a compressor that clamps down on a transient can cause the sustained resonant tail to ring more prominently. an EQ boost at 2 kHz can push a mild resonance into harshness territory.

multiple sources stacking

three tracks with mild 3 kHz peaks become a mix with a painful 3 kHz problem. resonances compound when sources are summed. this is why a mix that sounded fine on individual tracks can develop harshness or boom that was not there before.

finding resonances in your audio

before you can fix a resonance, you need to find it. there are three practical approaches:

the sweep test

insert a parametric EQ on the track. set a narrow band (Q around 8) with a 6-10 dB boost. slowly sweep it across the frequency range. when you hit a resonant frequency, it jumps out immediately: the ringing or harshness becomes unmistakable. note the frequency, then cut instead of boosting.

the spectrum analyzer

watch the spectrum while the track plays. resonant frequencies show up as peaks that persist while the surrounding content moves. they are narrow, consistent, and often visible as bright vertical lines in a spectrogram view.

the listen-and-compare test

A/B your track against a reference that sounds clean in the same frequency range. if your track has a resonance the reference does not, the difference is usually obvious once you know what to listen for.

the tools for controlling resonance

once you have identified a problem resonance, the approach depends on the nature of the problem:

consistent, fixed resonance (a room mode, a mic presence peak): a static EQ cut works. one narrow notch, set and forget. this is the simplest and cleanest solution.

intermittent resonance (harshness that comes and goes with dynamics): a dynamic EQ is better. it only cuts when the resonance exceeds a threshold, preserving the frequency’s natural contribution the rest of the time.

shifting, broadband resonance (harshness that moves across the spectrum with pitch or vowel changes): this is where dedicated resonance suppressors and spectral processing come in. they analyze the entire spectrum and act on every frequency simultaneously, following the resonance wherever it goes.[^2]

the later stages of this guide path cover each approach in detail. the point here is that understanding what resonance is and where it comes from lets you choose the right tool instead of reaching for the heaviest one every time.

frequently asked questions

frequently asked questions

what is resonance in audio production?

resonance is the tendency of certain frequencies to ring louder or longer than others. every physical system (a room, an instrument, a speaker cabinet) has natural resonant frequencies determined by its size and shape. when sound energy hits one of these frequencies, it builds up instead of passing through evenly. in a mix, this shows up as narrow frequency peaks that sound boomy, harsh, or ringing.

why does my vocal recording sound harsh?

vocal harshness usually comes from resonant build-up in the 2-5 kHz range, where your ears are most sensitive. common causes include the microphone presence peak (condenser mics boost this range by design), room reflections creating narrow peaks, and preamp or compressor behavior adding resonant coloring. the fix depends on whether the resonance is consistent or changes with the performance.

what is the difference between resonance and feedback?

feedback is a specific type of resonance that happens in a live sound system. when a microphone picks up its own amplified output, the signal loops and the resonant frequencies of the room and PA system build up into a loud, sustained tone. resonance in a studio or mix context is subtler: it is the natural emphasis of certain frequencies by the room, instrument, or signal chain.

how do I find resonant frequencies in my mix?

sweep a narrow EQ band (Q around 8-10) with a 6-10 dB boost across the frequency range. when you hit a resonant frequency, it will jump out: the ringing or harshness becomes obvious. you can also use a spectrum analyzer to look for narrow peaks that stay present regardless of what is playing. the most common problem areas are 200-400 Hz (room modes and boxiness) and 2-5 kHz (harshness and brittleness).

can resonance ever be a good thing?

absolutely. resonance is what gives instruments their character. a guitar body resonates to project sound. a singer uses chest and head resonance to shape their tone. a snare drum rings at its fundamental. the goal is never to eliminate all resonance. it is to control the resonances that hurt your mix while preserving the ones that give it life.

references

a note from the developer

this guide is built on four years of studying psychoacoustics and DSP research. reading papers, building prototypes, making mistakes, and learning from all of it. i am a solo developer in copenhagen, and i am still learning every day.

i started building KERN SMOOTH because i was tired of reaching for a broad EQ cut every time a vocal sounded harsh, knowing i was losing presence along with the harshness. the trade-off frustrated me, so i went looking for a better approach. what i found was ERB-scale spectral analysis: a way to see audio the way your ears actually hear it. 40 perceptual bands, each matched to how your auditory system resolves frequency. it took months to get it working transparently, and i am still refining it.

if i got something wrong, missed something important, or if you just want to share how you deal with resonance in your mixes, i genuinely want to hear from you. reach out at jonas@kernaudio.io. every piece of feedback makes these guides better.