EQ

What equalizing is

Equalizing means raising or lowering the volume of specific frequencies. As you saw in how sound works, a voice isn't a single tone but a mix of lows, mids and highs. With the equalizer (EQ) you decide how much of each zone you want: remove rumble, clean up the “mud” or add a bit more clarity. In the simulator above you can move three bands and a filter over the spectrum of a voice and hear it go from muddy to clear, or to harsh if you overdo it.

Each band of a parametric equalizer has three controls: frequency (where it acts), gain (how much it boosts or cuts, in dB) and Q or width (whether it affects a narrow or wide zone).

The map of the voice

These are the zones you'll touch most. The frequencies are approximate: every voice and every microphone is different, so use them as a starting point and always confirm by ear.

First, the high-pass filter

Almost every podcast voice benefits from a high-pass filter: it lets through what's above a frequency and removes what's below. Set between 80 and 100 Hz, it removes bumps on the desk, vibrations, traffic rumble and some of the plosives without touching the voice.

Set it carefully: with very deep voices, start at 70–80 Hz; with higher voices you can go up to 100–120 Hz. If you set it too high, the voice loses body and sounds thin, like on the phone.

Cut before you boost: subtractive EQ

The most common mistake is boosting everything that sounds good. It's far more effective to remove what bothers you: if the voice sounds dull, often it isn't missing highs but has too many low mids. Cutting the mud makes the presence appear on its own.

To find a problem frequency, use the sweep trick:

1. Create a band with a narrow Q and boost it a lot, about +8 dB.
2. Move it slowly between 200 and 600 Hz until the flaw (the “box”, the mud) sounds exaggerated.
3. That's the problem: now cut that band to −2 or −4 dB and widen it a little.

Presence, air and sibilance

Once the voice is clean, sometimes it lacks a little closeness. A gentle, wide boost in the presence zone, between 2 and 5 kHz, improves intelligibility, especially if the listener is on the street or in the car. A touch of air above 10 kHz adds brightness, but also raises the hiss, so do it after noise reduction.

When you boost the highs, watch for sibilance: S and Z sounds that hiss or pierce, between 5 and 8 kHz. That's what the de-esser is for, a compressor that acts only in that zone and only when an S appears. Set it to take off a few dB on the loudest S sounds; overdo it and the person will sound like they have a lisp.

Always compare with the original

The ear gets used to what it's hearing very quickly. After a few minutes of EQing, any change seems like an improvement. So you don't fool yourself:

• Switch the equalizer on and off often (bypass) to compare with the original.
• Compare at the same volume: whatever sounds louder always seems better.
• Listen on headphones and also on your phone's speaker, which is where a lot of people hear podcasts.
• Rest your ears and listen again the next day if you can.

If there are several voices, EQ each track separately: each person and each mic needs its own settings. The goal is for them to sound clear and consistent with each other, not identical. Compression comes next.

Your turn: find the mud in your voice

EQ a recording of yourself using only cuts and a minimal boost, and compare it with the original.

1. Put a high-pass filter at 80–100 Hz and listen to what disappears.
2. Sweep between 200 and 600 Hz, find the frequency that sounds worst and cut it by 2–4 dB.
3. Boost the presence by 1–2 dB at 3 kHz and compare with the original at the same volume, on headphones and on your phone.