Tips On Voice Equalization
The human voice is one of the most difficult audio sources to work with, here we discuss how to EQ it correctly.
The human voice is one of the most difficult audio sources to work with. This is mainly because every human voice is different. People’s different vocal chords and body shapes mean that no two voices are the same - this means that there are no “best equalizer settings” that can be applied to everyone. That being said, there are general rules that can be followed when dialling in your equalizer settings.
There are certain areas of the human voice frequency range that tend to have certain characteristics. Particular frequencies can add clarity and power to the human voice when boosted, whilst other ranges can reduce muddiness when cut.
Learning these different areas of the frequency spectrum can make the process of voice equalization a lot quicker and easier.
There are also certain frequencies that are prone to problems such as sibilance. EQs can be used to deal with these issues, however there are often more specialised tools better suited to dealing with them. We’ll have a closer look at the options available to you later on in this article.
However, as mentioned before, there are no definites when it comes to tweaking equalizer settings for voice - experiment and find what works best for you!
Let’s take a look at some tips you can use when working on voice equalization. As we’ve said many times these use the following advice as a starting point from which to experiment - each voice is different and these rules won’t work for everyone.
Cut out the sub frequencies
The human voice does not contain any sub frequencies (20 to 60 Hz) and so it’s normal practice to totally remove these frequencies with a high pass filter.
This is a good thing to do because there can sometimes be unwanted frequencies in this range that aren’t part of the vocal recording. These rogue sub frequencies are often present due to vibrations affecting the microphone which are picked up as bass frequencies.
The vibrations often occur as a result of the speaker moving their feet when recording their voice overs, but they can be present for a whole host of reasons.
To remove them, simply load an EQ on your voice recording and add a high pass filter at around 60Hz. Depending on the voice you may be able to cut frequencies all the way up to 80Hsz, experiment and see what sounds best! Be careful though, if you set the high pass filter too high you may lose power in your voice recording.
Identify the fundamental frequencies of the voice you’re working with
The fundamental frequency of the human voice (also known as f0) falls between 85 and 180 Hz for male voices and 165 to 255 Hz for female voices. Children’s voices tend to have a fundamental frequency at around 300 Hz.
On a visual EQ the fundamental frequencies of a voice can often be easily identified. Look for a peak in the fundamental frequency range - it will vary slightly with each syllable but will stay roughly at the same point in the frequency spectrum.
You can also use your ears to identify the fundamental frequency of a voice. Boost a bell band in the fundamental range and move it around until you find the frequency where the voice sounds most powerful.
How should you dial in your equalizer settings when working with fundamental voice frequencies? There are situations in which you may want to boost them and there are situations where you may want to cut them.
If your voice sounds quite thin, boosting the fundamental frequencies can thicken it up, giving a richer, fuller sound. However, too much boosting in this area can lead to boominess, so be careful.
If your voice recording already sounds quite boomy, cutting the fundamental frequencies should reduce this - just don’t cut too much! You’ll end up with a thin, weak sounding voice over.
Working with consonants and vowels
The vowels contain most of the power of the voice and so, if your voice recording is sounding a bit weak, it is often a good idea to boost them.
However, be careful. From 330-600Hz the voice can tend to sound a little “boxy'' so it’s best not to boost here. In fact, it’s often beneficial to cut in this area. Again, as with all the advice in this article, experiment. You may find that boxiness isn’t a problem with the voice you’re working with.
Consonants are responsible for intelligibility. Intelligibility is essential if you want your audience to engage with your content - no one’s going to stay and listen to what you have to say if they’re struggling to make out what you’re saying.
Boost frequencies in the 1.5KHz and 4KHz to improve the clarity of your voice recordings, try boosting different frequencies in this range to work out what works best for the voice you're working on.
Dealing with sibilance
Sibilance occurs when we say “ess” sounds. These consonants are high in energy and amplitude and are unpleasant and sometimes painful to listen to.
They occur in the frequency range of 4k-10kHz (again, exactly where they occur changes depending on the voice you’re working with). You can use an EQ VST to remove them, however, it’s better to use a specialised tool called a de-esser.
You only want to cut the sibilant frequencies when sibilance occurs. If you use an EQ then the sibilant frequencies will be cut all the time, even when there is no sibilance occurring. This can lead to a weaker, less clear voice over.
Getting the best equalizer settings can be a lot of work. If you want a quick and easy solution to voice equalization then you might want to try an automatic equalizer. Automatic equalizers (or auto EQs) listen to your audio source and apply an EQ curve for you.
This is a really easy solution if you find the world of audio intimidating, or just want to save time when editing your audio.