Join Sage Audio

How to Make Clean Vocals in 10 Steps

  • Start with De-clicking
  • Clean Vocal EQ
  • Use Split Band De-essing
  • Introduce Harmonics to the Vocal
  • Use a Subtle Vocal Rider
  • Utilize Clean Compression
  • Insert Upward Compressor
  • Attenuate Resonances with Soothe 2
  • Amplify Desired Frequencies
  • Introduce Volume Automation

Get a FREE Mastered Sample of Your Song ➜

Start with De-clicking

For this video let’s look at getting a vocal as spectrally and dynamically balanced as possible.

Let’s start by removing vocal clicks or the unwanted mouth sounds that a mic often picks up during a performance. To do this I can either use the RX De-Click plugin and set the algorithm to multi-band random click, or I can use Spiff by Oeksound.

Let’s use Spiff’s remove mouth clicks preset, and modify it slightly to better suit our vocal.

Watch video example

Clean Vocal EQ

Next, I want to use an EQ to remove aspects of the vocal that I want less of, or that make it sound unbalanced. I removed low-frequency rumble and hum with a 12dB/octave filter and dipped a little at 250 Hz. to reduce masking.

I found nasally tones right below 1kHz, and also some excessive sibilance at 13.1kHz. Let’s listen to these bands being introduced and notice how the vocal becomes slightly more balanced.

Watch video example

Use Split Band De-essing

Continuing our subtractive processing, let’s introduce a de-esser to attenuate high frequencies in a program-dependent way. I’m using a high shelf filter to attenuate 6.6kHz and above, and a bell filter to attenuate a small area where I found excessive sibilance - only by a couple of dB on each.

Like the EQ, this will help to balance the frequency spectrum of the vocal - let’s take a listen.

Watch video example

Introduce Harmonics to the Vocal

It may seem counter-intuitive to saturate a vocal if I want it to sound cleaner, but harmonics will improve the signal-to-noise ratio and make the vocal sound fuller. With this PSP saturator, I used subtle settings and a modern tape algorithm.

Additionally, I drove the input into the saturator a little, before compensating with the output.

Let’s listen and notice how the vocal sounds clearer, and slightly louder.

Watch video example

Use a Subtle Vocal Rider

Before I compress the vocal, there’s one more thing I want to do, which is to introduce a vocal rider - this plugin uses lookahead to alter the level of the vocal based on a pre-set target. As a result, the vocal’s dynamics will be more balanced.

This reduces the need for aggressive compression, which would impart a distinct, often unwanted timbre onto the vocal.

Let’s listen and notice how although the effect is subtle, the vocal is more controlled.

Watch video example

Utilize Clean Compression

When I want a clean-sounding vocal but need to compress, I like to use the FF-Pro-C2 and choose the clean algorithm, with a soft-knee and look-ahead. This reduces any coloration or timbre changes to the vocal, while still attenuating the signal as needed.

Oversampling will also help to keep a clean sound by reducing aliasing distortion. Let’s take a listen and notice how the compressor doesn’t change the sound, just the dynamics.

Watch video example

Insert Upward Compressor

An upward compressor will compress from the noise floor up - since we improved our signal-to-noise ratio in chapter 4, we can use this afterward to amplify quieter details without worrying too much about added noise. Also, it helps control dynamics, without the need for more peak-down processing.

Let’s listen and notice how the vocal sounds more detailed, dynamically controlled, and how the noise floor stays roughly the same due to the harmonics we introduced earlier.

Watch video example

Attenuate Resonances with Soothe 2

The original vocal contains resonances that were likely exacerbated by some of our processing - with that in mind, I like to save Soothe 2 for later on in the chain to ensure these resonances are attenuated. In this instance, I found the low mids to mids were the problem areas.

That said, I emphasized these ranges with the pre-eq, and attenuated. Let’s listen to how the processor improves the vocal, and also enable the delta function so we can hear what exactly is being removed.

Watch video example

Amplify Desired Frequencies

Last up in terms of processors, I like to add one more EQ to shape the overall sound. Some low rumble was reintroduced from our processing, so I’ve attenuated that with a high-pass, before lowering 250Hz. very slightly to again reduce masking in the higher frequencies.

I dipped some mids again, but then amplified 3.5kHz to increase the vocal’s presence, and included a shelf above 10kHz to introduce some air to the vocal.

Having an EQ at the very end of the chain gives you a lot of control over the final spectral balance of your vocal.

Let’s take a listen to how this clarifies the vocal.

Watch video example

Introduce Volume Automation

Last up let’s use volume automation - volume automation affects the fader, which routing-wise is placed after all of my inserts, now is a good time to introduce it. If you were to do this before inserting plugins, any changes might be negated or altered by the plugins.

With this automation, I can fine-tune the final volume of each section. Let’s listen to the final result, and keep in mind that this chain sets you up perfectly to add your final creative processors , like reverb, delay, chorusing, and whatever temporal processor you want to use.

If you have a mix that you need mastered, send it to us to receive a free mastered sample of it with the link in the description. Thank you so much for watching.

Watch video example

Get a FREE Mastered Sample of Your Song ➜

Join Sage Audio