How to EQ vocals in Ableton Live

Before you EQ a vocal

EQ can make a good recording sit better, but it can't rescue a bad one. If the vocal was recorded in a boxy room, too far from the mic, or with the gain so high it clipped, fix that first if you can. Re-recording one line usually takes less time than an hour of EQ, and it sounds better.

Even out the level before any processing. Vocals jump around a lot, with quiet breaths, soft words and loud held notes. In Live you can split the clip and set each part's gain in Clip View, or draw clip gain automation, so the phrases arrive at a similar level. That gives the EQ and the compressor a steadier signal to work on, and it means the compressor doesn't have to do all the evening out on its own.

Know what the de-esser is for. Sharp S and T sounds are a level problem in one narrow area, and they come and go with the words. A static EQ cut there dulls the whole vocal all the time to fix a problem that only happens on a few letters. A de-esser turns that area down only when an S or T hits. Live doesn't have a device called a de-esser, so the usual way is Compressor with its sidechain EQ focused on the sibilance. The free mix kit's Vocal Chain does exactly that, with its sidechain EQ set to 7 kHz and a De-ess knob to set how hard it works.

Put the EQ cuts before the de-esser and the compressor, so the compressor isn't reacting to mud. Boosts can go either side. A presence or air boost after the compressor stays exactly as you set it. The same boost before it, which is how the kit's Vocal Chain does it with one Channel EQ at the start, gets evened out a little by the compressor, and the de-esser has more top end to react to.

EQ in the full mix, not in solo

A vocal in solo always sounds like it needs more of something. More low end to sound warm, more top to sound clear. Make those moves and then bring the beat back in, and the vocal sounds thick and crowded, with the low end fighting the bass and keys.

Do the sweeping in solo if it helps you find a spot, then make every final decision with the whole track playing. The mud cut is the clearest example. In solo, cutting 3 dB in the low mids makes a vocal sound a little thin. In the mix, that same cut is what lets the words come forward without turning the vocal up.

Listen for the words, not the tone. The question isn't whether the vocal sounds rich, but whether you can understand every word at a normal listening level without straining. If you can, the vocal EQ is probably close. If a word gets lost behind a snare hit or a guitar strum, the fix may be in that other part, not in the vocal.

Rap, pop and backing vocals

The same bands apply to every vocal, but where you land in each one shifts with the style.

Rap vocals carry the rhythm and have to cut through a busy low end, so clarity matters more than air. A clean high pass and a firm mud cut help the most, and the presence area does the rest. Many rap vocals need little or no air shelf, especially over bright hi hats. A deep male voice may need the high pass a little lower so the chest of the voice stays.

Pop lead vocals usually sit bright and close. The presence boost and the air shelf matter more here, because the vocal is the main event and should sound polished. The catch is that a brighter vocal shows sibilance more, so the de-esser has more work to do.

Backing vocals should support the lead without competing with it. Cut more low end from them than from the lead, often with the high pass a good way higher, and take the presence down a little rather than up. That leaves the lead in front and lets the backing parts blend into a wider, softer layer behind it. Sending them to the same reverb return as the lead helps them sound like one group.

Mistakes that make vocals harsh

Boosting presence to fix a vocal that is just too quiet is the most common one. If the words get lost, try the fader first. A vocal that is loud enough needs far less boost to cut through, and a vocal that is too quiet will sound harsh long before it sounds clear.

Stacking boosts in the same area is another. A presence boost on the EQ, a bright reverb, a saturator adding top end and an air shelf can all pile up around the same few kilohertz. Each one sounds fine alone, and together they hurt. Check each stage by switching it off.

Narrow boosts can whistle. A boost with a high Q in the upper mids picks out one resonance and makes it ring. Keep boosts wide, and use narrow bands only for cuts.

Skipping the de-esser and cutting sibilance with a static EQ makes the vocal dull and still leaves the S sounds sharp. Treat the S sounds with dynamics, and leave the EQ for the tone.

Last, judging at high volume makes harshness easy to miss, because loud listening tires your ears. Check the vocal quietly. If it still sounds sharp at a low level, it will sound worse on earbuds.