It’s easy to ruin a good voice recording with heavy-handed equalization.
I’ve heard voice overs that sound like the speaker had a blocked nose—when they didn’t. I’ve heard voice overs that sound like the speaker was talking in a high-pitched voice—when they weren’t. I’ve even heard voice overs where the speaker sound so odd that the recording became simply unintelligible.
All of these voice recording started with a great sounding clean voice. But, unfortunately, they were butchered with aggressive, incorrectly applied EQ.
Yet you don’t need to be an experienced audio engineer to apply EQ that improves a voice recording, rather than ruining it.
In this tutorial, I'll show you how to approach equalization before going in to EQ specifically for voice.
It’s All Over Before You Know It
Before I start talking about EQ, I want to ensure you have the right frame of mind when it comes to recording voice.
The character and tone of any voice recording is decided long before you load up the DAW or audio editor and apply some EQ.
If you want your voice over to sound warm, you need to create this warmth in the recording phase. Too many people try to create warmth with EQ.
Unfortunately that’s not how it works. You can’t create new frequencies and completely change the character of a recording. You can only exaggerate or lessen what is already there. Sure, they can make a voice sound warmer, but the warmth needs to be there in the first place.
Before you load up an EQ to change the character of the voice, ...stop. If you don’t like the tone of the voice recording, you might have to start again.
Next time you record, think about what character and tone you want to capture. A~void thinking that you'll make it sound better with EQ and compression. Get it sounding good at the source, and you will have a far better final product.
How To Use An Equalizer
Equalizers are not to fix the character of a voice over but they can emphasise the good stuff. Or they can remove the bad stuff. It’s that simple.
They can be used to change the character of a voice, but only slightly. And they can be used to improve a voice recording, but only if it’s a good recording in the first place.
That’s the philosophy out of the way—now on to the practical application.
To start off with, you should only approach EQ in two ways. Later on you can start experimenting (once your ears are better trained). But for now, stick to these two approaches and you can't go far wrong.
- The first approach is to use narrow cuts to remove room resonances and unpleasant elements of the sound
- The second approach is to use wide boosts to exaggerate the good stuff
But here there are also two golden rules to consider at all times:
Try to use cuts more than boosts. This is called subtractive EQ. Without going into too much detail, it’s always better to cut. So if you want something to sound brighter, cut the lows instead of boosting the highs
Subtlety is key—especially when it comes to voice. Avoid cutting or boosting more than 3-5dB. Any more, and the voice will start to sound unnatural and odd
It’s also worth bearing in mind that these approaches only apply to single bands—bell curves—on the EQ. There are two other types of EQ—filters and shelves—that the golden rules don't apply to.
Shelves boost or cut everything below or above a certain frequency. A low shelf will boost/cut everything below it, and a high shelf will boost/cut everything above it.
The golden rules still apply to these: cut before boosting, and keep it subtle.
Filters completely cut everything below or above a certain frequency. A high pass filter will cut everything below—it lets the highs pass, and a low pass filter will cut everything above—it lets the lows pass.
The golden rules don’t apply to these: you can only cut with filters—there’s no option to boost—and you can’t limit them to 3 dB's—because they cut everything.
There we have it - three types of EQ, two approaches and two golden rules. Remember these and you can EQ anything. Just use your ears.
Applying EQ to a Voice Recording
Now, to consider EQ in relation to voice.
First, apply the general principles that I have just discussed. Move the EQ bands around until you find any room resonances, and cut them by 3dB. Move the bands around until you find pleasant elements of the sound, and boost them by 3dB.
No two voices are the same, so warmth in one voice will be in a slightly different place to another. Especially between males and females—male voices are centered around 80-180 Hz, whereas female voices are centered around 160-260 Hz. That's a big difference.
Having said that, there are a few areas that are common across all voices. Only use the following tips as guidelines and starting points—it’s important to experiment, use your ears and find what works for the particular voice.
Before I start, here's the voice recording with no processing:
1. Use a High Pass Filter to Cut Everything Below 80Hz
This is a common practice and something that you can do to improve any voice over. Anything below this frequency will be low end rumble and noise. Remove it, and it will instantly clean up your voice over.
Try going even higher, especially on a female voice. If your voice recording is sounding a bit too bass heavy, cutting everything below 100Hz will really help with intelligibility.
Here is the voice recording with a high pass filter applied at 80Hz:
2. Cut 100-300 Hz to Add Clarity
Similar to the last tip, cutting the bass will improve clarity. On the other hand, if the voice sounds a bit thin, try boosting somewhere in this frequency range.
Here is the voice recording with a cut from 100-300 Hz:
Here is the voice recording with a boost from 100-300 Hz:
3. Cut 300-400 Hz if the Voice Sounds ‘Muddy’
This is a problematic frequency range for most recordings—particularly in music. If the voice sounds a too muddy, try cutting somewhere around this area.
Here is the voice recording with a cut from 300-400 Hz:
4. A Wide Boost Between 2-6 kHz Can Improve Clarity
If cutting some of the bass around 100-300Hz doesn’t add enough clarity, try a gentle boost across this frequency range.
Here is the voice recording with a boost from 2-6 kHz:
Be careful of exaggerating the sibilance and S sounds of the voice though, which leads me on to…
5. Cut Around 3-5 kHz if the Voice Sounds too Sibilant
Be wary of this frequency range when boosting. Boosting too much at these frequencies can make the voice sound highly sibilant and add too much sizzle.
Here is the voice recording with a cut from 3-5 kHz:
Remember that these tips are purely intended to be used as starting points. Experiment, and trust your ears.
If you can’t hear the small changes that you’re making, try using the Altitude Mixing Technique. But remember that subtlety is key. Spend time getting the sound that you want when you record, and use EQ to improve it rather than fix it.
Have you ever had a bad experience with EQ? Or perhaps you have used EQ to make a good voice recording even better? Leave a comment below and tell me about your experiences.