EQ is arguably the most important signal processor for bringing out the best in vocal recordings. Learn in detail how EQ works, how to fix problems, and make your vocals sound even better.
By Craig Anderton
This is the fourth chapter of the “Recording Vocals at Home” series. Here you can find the full series.
Of all the processors in a chain of vocal effects, the equalizer is perhaps the most important. An equalizer amplifies (boosts) and/or attenuates (cuts) frequencies you specify, to change a vocal’s timbre. Equalization can make thin voices more full, muffled voices more articulated, and screechy voices sweeter.
The most common way to vary EQ is with plugins inserted in your DAW. But many factors influence your vocal’s frequency response, even before it reaches your audio interface:
- The mic itself. Mics sometimes boost the highs a bit. Condenser mics tend to be brighter than dynamic mics, but not always. Ribbon mics have a naturally rich bass.
- Mic switches. Some mics may have a bass rolloff switch, which gently reduces response in the bass range.
- Your proximity to the mic. With directional mics, singing closer to the mic increases the bass response.
- Mic position. Positioning the mic slightly below, above, or at an angle to your mouth can affect the tone.
- Room acoustics. This is a more subtle variation, but it can still affect the overall sound.
It’s worth doing a mic test, where you record yourself singing into the mic with words like “I’m about a foot away from the mic,” “I’m singing at an angle,” “the mic is pointing up at my mouth,” and so on. Listen back and see if one of these positions works best with your voice.
1. How Different Frequencies Affect Vocals
The bass range (around 100-200 Hz) gives fullness and depth. Examples of singers with deep voices include Barry White, Johnny Cash, Leonard Cohen and Nick Cave. However, your mic will also pick up frequencies below the vocal range, such as “pops” caused by blasts of air when singing plosives (e.g., the letters “p” or “b”).
Most of a male vocal’s energy is in the lower midrange of 100 to 350 Hz, while females are somewhat higher (e.g., 200 to 500 Hz). But the total vocal energy isn’t restricted to those ranges because the voice generates harmonics. Vowels typically produce energy further up in the midrange (up to around 1,000 Hz), while consonants are associated more with the upper midrange, around 2,000 to 5,000 Hz. The treble region above 5 kHz captures the sounds of breathiness and sibilants (like “s” sounds).
Note that these frequency ranges are approximate—a soprano or alto vocalist will cover more of the upper part of a range, while a tenor or bass singer will be more in the lower areas. However, they represent good starting points for adjusting EQ. For example, if the consonants don’t come through and the vocal lacks intelligibility, you’ll want to boost the frequencies where consonants are most prominent. Conversely, if the consonants are too prominent, you’ll probably want to reduce those frequencies.
2. Equalizer Parameters
There are four main equalizer parameters. However, not all filters have all parameters:
- Frequency sets where the boosting or cutting occurs in the audio spectrum.
- Boost/cut (peak/dip) determines the amount of amplification or attenuation at the selected frequency.
- Bandwidth, resonance or Q determines the sharpness of the boosting or cutting action. Narrow bandwidth settings affect a small part of the audio spectrum, while broad settings process a broader range.
- Slope applies only to highpass and lowpass filters (described next).
The decibel (dB) measures the ratio between the level of two audio signals, which we can use to specify the amount of boosting or cutting. A dB spec can also have a – or + sign. For example, cutting the low frequencies by -6 dB creates more attenuation than cutting by -3 dB. A setting of +2 dB would give a slight boost instead of a cut, while a setting of +10 dB would create a major boost.
In addition to these controls, use the bypass button to compare the unequalized and equalized sounds as a reality check. Just a few dBs of change can make a big difference. Also, avoid “iterative” EQ re-tweaking where the lows seem thin, so you boost the bass, but now the highs don’t seem clear, so you boost the highs, and so on. Focus on the frequency range with the biggest problem, and fix that before moving on to any other frequency ranges. For example, if the vocal sounds muddy, reducing the lower midrange or bass response may solve the problem without needing to tweak anything else.
3. Equalizer Responses
Equalizers often include multiple stages with different filter responses, so you can reject some frequencies while boosting others.
Many of the screenshots show the Waves F6, which is the company’s flagship equalizer. It combines elements of equalization and dynamics. For this installment of the series, we’re only using the equalization aspects. The next installment covers dynamics control, and we’ll refer back to the F6 to describe how it combines this functionality with equalization. In the screenshots, controls that don’t relate to what’s being illustrated are grayed out, so you can focus on the controls that relate solely to equalization.
The highpass and lowpass filters attenuate audio below and above their cutoff frequencies, respectively. The cutoff frequency is where the attenuation starts becoming significant (Fig. 1). The slope sets the rate at which the attenuation occurs past the cutoff frequency. The steeper the slope, the greater the attenuation as you move further away from the cutoff frequency.
Highpass filters are often used with vocals to attenuate sounds below the vocal’s note range, like mic handling noises, plosives, room rumble, and the like. Lowpass filters are less commonly used with vocals.
A shelf response (Fig. 2) starts boosting or cutting the highs starting at a particular frequency, then levels off to a constant amount of boost or cut.
Shelving EQ is excellent for general, gentle tone-shaping. For example, if a vocal isn’t bright enough, a treble-boost shelf may be the answer. Attenuating the bass with a shelf can tighten up the sound while boosting the bass can add depth to thin voices.
The Peak/Dip or parametric response (Fig. 3) boosts or cuts only those frequencies around its resonant frequency. The Q parameter that selects the range of frequencies affected by the peak or dip is called the bandwidth. Peak is also called bandpass or bell, while dip is also called band reject or notch.
Parametric equalizers are used mainly for problem-solving and more detailed equalization applications. For example, a low-frequency shelf can reduce all the lows below a certain frequency, but a parametric can be more selective about which low frequencies to reduce. Another use is for reducing resonances that are louder than desired or boosting frequencies that seem weak by comparison.
4. Hear EQ in Practice
Let’s show how equalization can solve common problems. Before doing that, though, it’s time for a shoutout to Martha Davis (who had a string of hits in the ‘80s with The Motels and maintains an active recording and touring schedule), who agreed to record some vocal demos for these examples. The only stipulation was that she could do only one take, had to use cheap gear, and couldn’t do anything to improve the vocal. Well, even under those conditions, her voice was stellar, but I did manage to find a couple of problems. First, we’ll use the F6 to reduce unwanted resonance.
- Example 1 – Remove Resonance
The phrase “It’s like I told you, only the lonely can play” plays twice. The first time, note the resonance on “like I,” and how in the second time, the resonance is gone. Fig. 4 shows how it was done.
In the top view of the unprocessed vocal, the resonance is highlighted in orange. In the bottom view, note how EQ Stage 4 has attenuated the signal sharply and deeply at that frequency. However, the cut around 2 kHz dulled the vocal somewhat. So, Stage 6 creates a slight, gentle boost around 4 kHz. It’s out of the range of the resonance and adds a little brightness.
Another common vocal problem is excessive sibilance sounds, like “s.” This is such a common problem that Waves created Sibilance, a special-purpose equalizer that reduces frequencies in the sibilance range—but only when a strong sibilance occurs. We’ll use Sibilance to reduce aggressive “s” sounds.
- Example 2 – Reduce Sibilance
The first time the phrase “We smiled, without any style…we kiss altogether wrong” plays, the “s” in “style” is a tiny bit too prominent, but the “ss” in “kiss” is like an unwelcome guest. When the phrase repeats, Sibilance has brought the “ss” in “kiss” in line with the rest of the vocal. Fig. 5 shows the setting that solved the problem.
Now let’s look at an example of salvaging a vocal with EQ, this time with my voice. The context is that when songwriting, I sit at my computer with an inexpensive dynamic mic and record a scratch vocal. Later, I use a ribbon or condenser mic for the real vocal. However, in this case, there was one phrase in the scratch vocal that I preferred, but the sound quality was…well, horrible. Here’s proof.
- Example 3 – Moonlight EQ Fix
The first time the phrase plays, it sounds muffled, indistinct, and the level of the bass in the latter part of the phrase is overpowering. The second time it plays, all these issues are fixed. Also, note that I added some reverb, so you could hear how the excessive low frequencies made the reverb boomy, but how removing them tightened the reverb sound. Fig. 6 shows the EQ settings.
The highpass filter brings the bass level into line. To fix the muffled sound, Stage 6 adds a high-shelf for an overall treble boost. Stage 3 adds a little lift at 2 kHz to increase intelligibility.
The example shows why it’s crucial to know how an EQ’s controls work rather than depend on presets. Try as I might, I couldn’t find a preset anywhere called “Fix Craig’s Scratch Vocal on Under the Moonlight.”
Finally, let’s deal with one of the banes of vocalists, narrators and podcasters—pops from plosives. To make sure the point got across, I sang the phrase “pick a pack of peppers.” When the phrase plays the first time, even with a windscreen, the pops are bothersome (Fig. 7).
- Example 4 – Pop Reduction
When the phrase plays the second time, the bad part of the pops is reduced. Again, note how changing the EQ changed the reverb’s effect, even though there were no changes to the reverb’s parameters.
Although some pops are still somewhat present, remember that no rational person is going to sing a phrase like “pick a pack of peppers.” You’ll have an occasional pop, which the music will cover-up after you get rid of the most egregious bass frequencies. For this example, I used the Waves Renaissance EQ (Fig. 8), although, of course, the F6 could do this as well.
The left side of Fig. 8 shows the unaltered sound. In the background graph, note the extreme amount of low-frequency content and how it’s gone in the image on the right. The filtering uses all of the EQ’s six stages—Stage 1 is a highpass filter to get rid of as much bass as possible, and Stage 2 adds a shelf to take the lows down even further. Finally, Stages 3-6 pile on with parametric cuts in the bass range to reduce the lows as much as possible.
5. Vocal Tips for Recording & Mixing
Now that we know what kind of tools we can use, let’s cover some general tips about mixing with voice.
There are two components to the highs. One is the upper midrange, which provides intelligibility; the treble range starting at about 5 – 6 kHz gives “air” and transparency. A high-frequency shelf with little or no resonance often works well unless the vocal is hissy or there are “ess” problems that a de-esser can’t fix. In that case, the extended response above 8 kHz or so might add noise and may not help the vocal anyway. Apply a parametric boost instead, with a wide Q, in the 4 – 7 kHz range. This should give a glossy, intelligible high-frequency response without boosting the ultra-high frequencies.
Listen carefully to the vocal in context with the rest of the mix because fixing the lows and highs may be all that’s needed. Set the vocal’s level in relation to the mix so that you can hear the low and high frequencies clearly. But if the vocal still sits too far back in a busy track, focus on the upper mids with a stage of parametric boost, typically in the 2.5 – 4.5 kHz range. It’s not necessarily a problem if this overlaps with a high-frequency shelf that provides a general boost because the upper-mid EQ provides a more focused boost.
The human ear is most sensitive in this frequency range. Use a slight boost with low-to-moderate Q and sweep slowly across the upper mids. There will usually be a frequency where the vocal sounds are present and intelligible. Avoid too much boost because too much emphasis in this range can sound harsh, and also, this may make the lows and highs seem deficient. If the vocal still doesn’t seem prominent enough after a conservative upper midrange boost, then you probably need to raise the vocal’s overall level.
Lower midrange/upper bass
One final problem may be too much energy, around 300 – 400 Hz. Because many instruments produce energy in this range, the sounds can pile up and get muddy. A slight, somewhat broad cut in this area can tighten up the vocal.
6. Analog EQ Flavors
EQs have different design goals. The F6 and Renaissance EQs mentioned here are intended to be useful, universal and flexible. However, Waves also makes models of vintage analog EQs, which are often designed to impart a particular sonic character. The H-EQ Hybrid Equalizer is a good example of this kind of design.
On the other hand, some of these vintage EQs can streamline sessions because they limit the control settings to the equivalent of “EQs greatest hits”—like many of the hardware EQs used in older recordings. Because they were analog, they couldn’t offer the flexibility of today’s digital processors. So, their options were refined over the years to make sure that the available settings provided the most benefit. The PuigTec EQs and the V-EQ3 fall under this category.
Furthermore, there are Signature Series plugins designed in conjunction with well-known engineers and producers. These are often designed to be complete processors for instruments like guitar, vocals, etc. So, if you’re using a plugin like the CLA Vocals, you know that the EQ capabilities have been optimized specifically for voice.
But no matter what kind of EQ you use, above all, listen. Vocals are the most important element in any song, and they need to be intelligible, powerful when needed, or intimate if appropriate. EQ can be a huge factor in how well your vocals come across. If something doesn’t sound right, set a slight amount of cut, and sweep the frequency back and forth. Note when reducing the level at a particular frequency improves the sound. Then, with another stage, set a slight amount of boost, and again, sweep the frequency back and forth to see if there are any “sweet spots” that give the vocal better sound quality. Although your initial choices may not end up being your final choices, they can give you some important clues about how you need to EQ your vocal for maximum effectiveness.
This is the fourth chapter of the “Recording Vocals at Home” series. Here you can read chapter 1 on mic choice and technique.
Musician/author Craig Anderton is an internationally recognized authority on music and technology. He has played on, produced, or mastered over 20 major label recordings and hundreds of tracks, authored 45 books, toured extensively during the 60s, played Carnegie Hall, worked as a studio musician in the 70s, written over a thousand articles, lectured on technology and the arts (in 10 countries, 38 U.S. states, and three languages), and done sound design and consulting work for numerous music industry companies. He is the current President of the MIDI Association. www.craiganderton.org.
Want to get more tips straight to your inbox? Subscribe to our newsletter here.