Automatic equalisation: how it works and when to use it
Last updated 6 min readTags:MixingEQEqualizationAutomatic EQ
Learn how automatic EQ works, when it is useful and how to check its changes. Includes reference matching, resonance control and room correction.
Automatic equalisation analyses audio and sets EQ adjustments for you. Depending on the tool, it may match a reference recording, work towards a chosen tonal profile or reduce prominent resonances.
It can provide a useful first pass. Before keeping the changes, compare them at the same loudness and listen within the full mix. A closer match on a graph does not necessarily sound better.
How automatic EQ works
With manual EQ, you choose which frequencies to boost or cut, by how much and over what bandwidth. Automatic EQ makes some of those decisions from its analysis. You may still need to choose a profile, supply a reference or set the processing strength.
Tools differ in what they analyse and how they respond. These functions can overlap within one plug-in.
Reference matching
A matching EQ compares the average frequency spectrum of your audio with a reference. It then creates a curve to reduce the differences.
For example, a mix with more bass and less treble than its reference may receive a broad bass cut and treble boost. Apple’s Match EQ includes controls for both the amount of processing and the smoothness of the curve.[1]
Choose a reference with a sound you want and a comparable arrangement. Compare similar sections, such as two full choruses. A sparse verse and a busy chorus may have different spectra for perfectly good musical reasons.
Profile-based processing
Some tools use profiles for vocals, instruments or complete mixes. These provide a target or guide for the processing. Some profiles use machine learning; others use predefined curves or rules.
A profile gives the tool a starting assumption about the sound you want. It does not establish the correct balance for every recording. Check what it changes, particularly if the source has a deliberately dark, bright or bass-heavy character.
Resonance reduction
Resonance-control tools reduce prominent frequency peaks. Some adjust their attenuation as those peaks change over time.
This can be useful for an intrusive ringing or harsh sound. However, a peak may also be a musical note or part of an instrument’s character. Listen for a specific improvement rather than trying to remove every peak on the display.
Automatic EQ and dynamic EQ are related but distinct. “Automatic” describes how the settings are chosen. “Dynamic” describes EQ whose gain changes in response to the signal. A tool may do either or both.
When automatic EQ is useful
| Task | What automatic EQ can offer | What to check |
|---|---|---|
| Bring a mix closer to a reference | A proposed change to the overall frequency balance | Whether the reference suits the arrangement and intended sound |
| Make recordings sound more consistent | A closer tonal match between takes or microphones | Whether changes in performance or microphone position need separate attention |
| Reduce an intrusive resonance | Attenuation around prominent peaks | Whether the processing also removes body or makes notes sound uneven |
| Reduce masking between instruments | Adjustments based on several tracks, where the tool supports this | Whether the intended lead instrument remains clear |
| Explore a starting balance | A quick alternative to compare with your current settings | Whether it improves the sound once loudness is matched |
The important questions are what the tool can analyse and what you want to change. A plug-in receiving only a vocal has no direct information about separate guitar tracks. A tool with linked instances or a sidechain may have more context.
For example, sonible smart:EQ 4 supports cross-channel processing between grouped tracks. Its group mode addresses masking between those tracks.[2] Automatic processing can therefore assist with relationships within an arrangement, though you must still decide which instruments should take priority.
When to adjust or reject the result
The reference does not suit the material
A reference with a prominent bass part may suggest more low end than your song needs. A bright acoustic recording may be a poor target for a deliberately dark electric guitar.
Try another reference, restrict the affected frequency range or reduce the processing strength. Do not force the whole spectrum to match merely because the control permits it.
The correction removes something you like
An automatic curve may reduce the weight of a vocal or the bite of a guitar. If the original character serves the song, keep it.
Detailed matching can also produce excessive boosts and cuts. iZotope’s matching EQ documentation warns that an exact match can produce unnatural results, and advises a gentler amount and more smoothing.[3]
The problem needs a different adjustment
A quiet vocal may need a fader change or automation. A crowded chorus may need fewer competing parts. EQ is one possible response, and an automatic suggestion is worth comparing with those simpler changes.
If a problem affects only one instrument, try correcting that track before applying EQ across the stereo mix. See master equalisation for adjustments to a complete mix.
How to check an automatic EQ setting
- Name the problem first. For example: “The vocal becomes harsh on loud notes” or “The mix sounds too bass-heavy beside this reference.”
- Analyse a representative passage. Include the section where the problem occurs. If the sound changes substantially elsewhere, check those sections too.
- Start with restrained processing. Reduce the amount or smooth a detailed matching curve. Limit the frequency range if only part of the spectrum needs attention.
- Match the output loudness. Adjust output gain so that the processed and bypassed versions sound equally loud. Extra level can make a change seem more impressive.
- Compare within the mix. Listen for the original problem, then check for losses elsewhere. A clearer vocal is little use if it has become thin or unpleasant.
- Keep only useful changes. Reduce, edit or bypass the result if it does not improve the mix. There is no need to retain the full suggested curve.
For a vocal, listen for clearer words without excessive sharpness or lost body. For a complete mix, check the weight of the kick and bass, the vocal level and the cymbal balance.
Is automatic room correction the same thing?
Room correction measures a playback system, usually with a measurement microphone, and adjusts its response towards a target. It is a different task from applying automatic EQ to a recorded vocal or stereo mix.
EQ can reduce some response peaks, but it cannot solve every acoustic problem. Genelec notes that EQ cannot correct cancellation caused by certain wall reflections; speaker placement or acoustic treatment may be needed.[4]
Keep monitoring correction in the playback path, outside the audio you export. On a recording, EQ may reduce an audible tonal problem, but it does not remove the room’s reverberation.
Where Sonalix fits
Sonalix analyses your mix and suggests areas to check. You make the changes in your DAW; it does not apply an automatic EQ curve to the audio.
Use its frequency-balance findings to guide your listening. A prominent frequency region is a reason to investigate, rather than an instruction to cut it. After a change, compare the sound at matched loudness and check whether the original problem has improved.