Skip to content

Frequency response curve: how to read your mix’s spectrum

Send feedback

Last updated 4 min readTags:MixingFrequencyAnalysis

Understand your mix’s frequency response curve, why it uses a logarithmic scale, and how to assess peaks, dips and roll-offs before you reach for EQ.

Sonalix’s frequency response curve shows the average level of frequencies in your mix, from bass to treble. Use it to investigate what you hear: an overpowering low end, a congested midrange or a lack of brightness.

The curve can suggest where to listen more closely. It cannot tell you, by itself, whether the mix sounds right.

What does the curve measure?

Sonalix displays 100 points between 20 Hz and 20 kHz. Each point represents a frequency and its average spectral level across the analysed audio.

Strictly speaking, this is an averaged frequency spectrum. A frequency response measurement describes how a device or system changes a signal at different frequencies. This chart describes the frequency content of your audio file.

The horizontal axis shows frequency in hertz (Hz). Bass sits on the left, midrange in the middle and treble on the right. The vertical axis shows spectral level in decibels (dB). Higher points indicate stronger content at those frequencies.

These values are not loudness readings. They do not tell you how loud each frequency sounds to a listener.

Why the frequency axis is logarithmic

A logarithmic axis gives each octave the same width. An octave is a doubling of frequency: 100 Hz to 200 Hz is one octave, as is 1 kHz to 2 kHz.

On a linear chart from 0 Hz to 20 kHz, everything below 200 Hz occupies just 1% of the width. The range from 10 kHz to 20 kHz takes up half the chart, although it covers only one octave.

Logarithmic spacing gives the bass more room on the page. You can inspect low-frequency changes without squeezing several octaves into the left-hand edge.

This spacing reflects musical pitch intervals. It does not mean that every octave sounds equally loud, or that every visible change needs correction.

How to interpret peaks, dips and slopes

Start with something you can hear, then inspect the relevant part of the curve.

What you notice What to check by ear
Strong low-frequency content Does the kick or bass overpower the rest of the mix, or is that weight intentional?
A broad rise around 200–400 Hz Does the mix sound congested, or does this region provide useful body?
Prominent upper mids Do vocals, guitars or percussion sound harsh, or appropriately forward?
A pronounced high-frequency roll-off Does the mix lack brightness, or does the arrangement suit a softer sound?

A peak may come from a sustained note, an instrument’s harmonics or a resonance. The curve alone cannot distinguish between them. A dip may reflect the arrangement rather than missing content.

A balanced mix does not need a flat curve. A spectrum that falls towards the treble can be entirely normal. Its apparent slope also depends on how the analyser scales and processes the data. Do not use the angle of the line as a quality score.

How Sonalix builds the curve

Sonalix analyses a mono downmix in short, overlapping sections using a short-time Fourier transform (STFT). It averages the magnitude spectra across those sections, converts the result to decibels and selects values at 100 logarithmically spaced frequencies.

The result is an overview of the analysed audio. It does not show how the spectrum changes from moment to moment.

What the curve cannot tell you

Brief problems may disappear into the average. A harsh cymbal hit or one boomy bass note may be much less obvious than a tonal imbalance throughout the track. Return to the relevant passage in your DAW.

The display has limited detail. The 100 plotted points do not represent 100 independently resolved measurements. Several low-frequency points may use the same underlying analysis bin. Logarithmic spacing improves readability; it does not add measurement detail. Use a higher-resolution spectrum analyser for narrow resonances or precise EQ decisions.

It cannot identify the responsible instrument. A low-mid rise could involve bass, guitars, keys or several parts together. Check the individual tracks before applying EQ to the whole mix.

A mono downmix can change the spectrum. Out-of-phase content in the left and right channels may partly cancel when combined. If something sounds weaker in mono, check the stereo relationship before boosting it.

How to use it during a mix revision

  1. Listen first. Describe the audible issue, such as a bass part that overwhelms the vocal.
  2. Inspect the relevant region. Look for a broad feature that might support what you hear.
  3. Find the cause in your DAW. Check the instruments and passages involved. The answer may be a level change, arrangement edit or automation rather than EQ.
  4. Compare the revision at a matched playback level. Keep the change if it improves the sound. A smoother-looking curve is not sufficient evidence.

The five-band energy view gives a broader summary of the spectrum. Use the curve when you want to inspect changes within those regions. Neither view supplies a universal target shape for a good mix.

Was this page helpful?

    Type to search titles, headings and summaries.