What is an EQ?

The short answer

An EQ, short for equalizer, is a tool that raises or lowers the volume of specific frequency ranges within a single sound. Instead of turning the whole signal up or down, you pick a region, say the deep lows or the crisp highs, and adjust only that region. Turning a region up is a boost. Turning it down is a cut. That is the entire idea. Everything else is just how precisely you aim.

Every sound is a blend of frequencies measured in hertz (Hz), from the low rumble of a room around 60 Hz up to the airy sparkle above 10 kHz. A voice, a guitar, and a kick drum each occupy their own mix of those frequencies. An EQ lets you sculpt that mix so a sound is clearer, warmer, or brighter without re-recording anything.

Bands, frequency, gain, and Q

A parametric EQ is the most common kind, and it gives you a handful of adjustable points called bands. Each band is defined by three controls, and once you understand them the whole tool clicks into place.

• Frequency. Where the band sits, in hertz. This is the center of the region you are about to change, from deep lows to bright highs.
• Gain. How much you boost or cut at that frequency, measured in decibels (dB). Positive gain lifts the region, negative gain scoops it out.
• Q. How wide or narrow the band is. A low Q affects a broad stretch of nearby frequencies for gentle, musical shaping. A high Q is a surgical spike that touches only a sliver, useful for hunting down one specific problem tone.

Read the curve left to right as low frequencies to high frequencies, and bottom to top as quieter to louder. A dip means that region is being cut; a peak means it is being boosted. When people talk about a "smiley face" EQ or a "notch," they are just describing the shape of this curve.

Why you EQ a voice

Voices are where EQ pays off fastest, because a microphone rarely captures a voice the way you want it to sound. Three moves cover most situations, and each maps directly to a band on a parametric EQ.

• Remove rumble. Cut the deep lows, roughly below 80 Hz, to strip out desk thumps, footsteps, and the low-frequency hum a room adds. Your voice does not live down there, so nothing you care about is lost.
• Tame boxiness. A gentle cut in the low mids, often around 250 to 500 Hz, clears the muddy, cardboard-box quality that close micing and small rooms tend to pile on.
• Add presence. A modest boost in the upper mids and highs, around 3 to 6 kHz, brings clarity and intelligibility forward so your words cut through a mix without needing more overall volume.

EQ is one piece of a clean voice chain. It works best alongside a noise gate to silence the gaps and a compressor to even out level. Getting the room quiet first matters too; if you are fighting a noisy background, see how to reduce background noise on your mic before you reach for heavy EQ moves.

EQ is a DSP effect

Under the hood, an EQ is a piece of digital signal processing. Your audio arrives as a stream of numbers, and the EQ runs math on that stream in real time to raise or lower the frequencies you targeted, then passes the result along. Because it is math rather than a physical knob on a circuit, a software EQ can offer as many bands as you like and recall your exact settings instantly.

This is also why EQ is cheap to run. A well-built parametric EQ adds a vanishingly small amount of processing time, which is why it sits comfortably inside a live signal path without you ever noticing a delay.

The parametric EQ in patchd

In patchd, a Windows virtual audio mixer, a parametric EQ is a free node in the effects rack on every channel. Drop it onto your microphone channel and you have frequency, gain, and Q controls for each band, with the curve drawn live as you adjust it.

The whole rack is yours to arrange. Each channel carries a noise gate, EQ, compressor, de-esser, AI noise suppression, saturator, and more, and you reorder them by dragging. A common voice order is gate first to cut the silence, then EQ to shape the tone, then a compressor to steady the level. The engine is real time, so shaping happens as you speak. Running through a bus at a small ASIO buffer, patchd adds only single-digit milliseconds of latency, about 10.7 ms at a 512 buffer, low enough to talk and monitor yourself live.

A bus in patchd is a virtual output that other apps pick up as a microphone, so the EQ-shaped voice you build here is what your call app, stream, or recorder actually receives. Routing is color coded to keep it readable: hardware outputs in green, amber, and coral, and buses in cyan, violet, and magenta.

Getting started

You do not need to memorize frequencies to benefit from EQ. Start with one band as a high cut to remove rumble, a second as a small dip to clear boxiness, and a third as a gentle presence lift, then trust your ears and adjust. Small moves of a few decibels go a long way; if you find yourself boosting more than about 6 dB, the recording or the room is usually the real thing to fix.

patchd is pre-launch, and its free effects rack, EQ included, ships on every channel from day one. If you want a clean, shapeable voice chain on Windows, join the waitlist and get notified when it ships. The paid Studio tier is $39.99 per year, and the parametric EQ is free either way.