When learning how to mix music, beginner engineers can often become discouraged when comparing their work to professional mixes. They know the tools (EQ, compressor, etc.) and how to use them, but for some reason they don’t get the same results. However, knowing what’s actually happening in the sound of a professional mix can help clear things up. In no domain is this more obvious and important than in mixing vocals.
In this article, we’ll discuss what makes a professional vocal sound...well...professional. We’ll cover frequency content, relative loudness, compression, vocal riding, doubling, and how Nectar 3’s Vocal Assistant can help bring your vocal to that professional grade.
Note that the characteristics we’ll be referencing are most obviously found in popular music. Vocals can be processed in an infinite number of creative ways, even in pop music, but these standard characteristics are tried and true across a variety of genres.
Vocal Assistant is one of the assistive audio technologies found in iZotope plug-ins, similar to Neutron’s Track Assistant and Ozone’s Master Assistant. Vocal Assistant is able to analyze a vocal and identify what processing is needed based on a few broad characteristics set before analysis.
Nectar 3’s Vocal Assistant includes two features: Assist and Unmask. We covered the Unmask feature a bit in “Making Space for Vocals,” which can help to create space in the mix so that the vocal is not competing with other elements. We’ll touch on Assist in this article, which is more helpful for setting an initial processing chain to make our vocal sound more professional.
After opening the Vocal Assistant, we can select the Assist feature to get started.
Next, we’re able to instruct Vocal Assistant to apply processing according to our needs. The “Vibes” parameters tell Vocal Assistant the sound that we’re going for, while the “Intensity” parameters indicate how much processing we would like Vocal Assistant to perform.
With these two parameters set, Nectar 3 will analyze playback. Keeping in mind the characteristics of a professional-sounding vocal, Vocal Assistant will load processing modules (EQs, compressors, etc.) in Nectar 3 to polish even a rough vocal.
As you would expect, all parameters on these modules can then be tweaked to fit your needs. Utilizing the Vocal Assistant is a great way to reach a starting point for mixing vocals, and in many cases may be all you need to get that professional sound.
Now that we’ve seen how helpful Vocal Assistant can be, let’s delve a bit further into each of the characteristics that make a vocal sound professional.
When analyzing a professional vocal, there are a couple areas of the frequency spectrum that should immediately stand out.
First, and most importantly, is the upper end of the spectrum. Professional vocals are noticeably bright, with a bit of hiss in the high-end. Listen to any song on the charts and you’re likely to notice this brightness. Sibilant sounds (s’s, ch’s, sh’s, etc.) are still controlled with de-essing, but the vocal as a whole has a high-end fizz that sounds crisp and pleasing.
Why is this the case? First of all, this brightness is extremely helpful for intelligibility. While some songs use the vocal as more of a support layer, the vocal is often the main focal element. And what good is that main element if you can’t understand the lyrics?
Brightness in a vocal allows the listener to understand more of the lyrics, as it allows the vocal to have presence and cut through the mix. In fact, this can often be the correct fix when the vocal sounds “too quiet.” Instead of turning the vocal up a few dB, simply use a high shelf filter to boost some of the high frequencies, usually above 5–10 kHz.
Next, notice the lows and low-mids in a professional vocal. Even with a female vocal, which usually starts at 200–400 Hz, there is some weight in these lower frequencies.
This is because the low and low-mid frequencies allow the vocal to have more prominence in the mix. The vocal will sound more intimate, like it’s closer to the listener, if some lower frequencies are allowed to come through. It’s a common mistake to cut too much low-frequency content out of a vocal, and doing so can cause a vocal to come across as wispy and powerless.
There’s an easy way to make sure your high pass filter is set to have the right cutoff frequency. Find the lowest note sung by the vocalist in your track. Your HPF should not be set any higher than this low note. If it’s set higher, certain notes will receive different filtering than others, with lower notes being filtered more than higher notes.
It can even be beneficial to set your HPF a bit lower or to use a lower pole number, as this will allow a bit of organic low frequencies to cut through. This will make the vocal itself sound more organic and give it enough low frequency content for it to have the power it needs.
Another characteristic of a professional vocal is its level in the overall mix. There is an increasing and prevalent trend in popular music of the super loud vocal, in which the vocal is considerably louder than other elements in the mix.
This naturally makes sense, as popular music tends to rely heavily on the vocal. Additionally, we see a higher percentage of solo vocalists as artists in the music industry today, so the importance placed on the vocal in a mixing perspective is expected.
The best way to achieve the ideal level for your vocal in the mix is to utilize reference tracks. Find a track that has a similar style to yours and pay attention to the vocal level. It’s a good idea to use the kick, snare, and most audible instrument as your reference points for how loud the vocal is in the mix. Compare to your mix and make adjustments as necessary.
Next, let’s discuss compression. As is the same with music in general, vocals have become progressively more compressed over the last few decades. Listen to most charting music and you’ll notice that there is barely a difference in loudness between whispered sections and belted ones. This applies to nearly every genre, from pop to hip-hop to electronic music.
However, if you try to compress your vocal as hard as the vocals in popular music, you may hear your compressor audibly working. Professional vocals sound compressed, but the compression is often extremely transparent.
This is possible in part due to vocal riding. Vocal riding is the process of recording volume automation with a fader while the vocal is playing, reacting to peaks and troughs in the vocal performance to smooth out level differences through the performance. This technique has been used since the days of analog tape, and even today should not be forgotten.
For one, volume automation is much more transparent than compression. The fact that a compressor has an attack and release time can result in compression being heard, rather than just affecting the dynamics of the vocal.
Additionally, riding the volume of a vocal will smooth it out so that a compressor won’t have to work as hard afterward. Having the main peaks and troughs smoothed out will allow you to compress more conservatively, resulting in a consistent vocal level characteristic of professional vocals.
Nectar 3 has the ability to perform what is essentially automatic vocal riding using the Auto Level Mode, located here:
Using the Vocal Assistant, Nectar 3 can set an optimized input level. The user is then able to set the target level of the Auto Level mode so that the vocal is normalized to that level. Nectar 3 recognizes dips in level and is able to adjust the vocal level to be more consistent over the course of the performance. Check out this video to see the Auto Level Mode in action.
Another important aspect of the professional vocal is doubling. This is the process of layering the main vocal with other takes of the same part, which are at a lower volume than the main layer. This can be done with toplines and background vocals alike and is done to add thickness to a vocal.
A couple plug-ins can help us to perform this doubling automatically. First is iZotope’s free Vocal Doubler plug-in.
This can take a single vocal layer and create a doubled effect, creating thickness and giving the vocal a sense of space in the stereo field. The user is also able to adjust the separation between the layers and add an element of organic human variation to make the doubling sound more natural.
The doubled layer can also be solo’d, allowing the user to fine-tune it. The vocal could be copied to a second channel or sent to a return track containing Vocal Doubler so that the original and added layers can be affected independently.
Check out this video to see Vocal Doubler in action.
This is regularly used to harmonize a vocal layer by introducing transposed copies of the original layer. We can use this feature for doubling by setting each voice to play in unison with the original layer, effectively adding copies at the same pitch. We can adjust pitch and time variation to once again give the doubling a more natural and humanlike feel.
Of course, it’s also possible to perform doubling manually.
Keep in mind that the added layers cannot just be copied and pasted from the original. This would result in perfect in-phase constructive interference, which would simply increase the amplitude and just make the vocal louder. Ideally, these added layers are just new takes, which the vocalist records while listening to the layer that they’re doubling.
It’s important to have these doubled takes lined up as closely as possible with the original layer. The point is to make the main vocal layer sound thicker, not to have two or more vocals audible at the same time. This is the reason that the doubled layers are often at a lower volume than the main.
As we just mentioned, the doubled layers are ideally recorded while the vocalist is listening to the original layer. Lining up doubled takes after the fact can cause the doubles to sound unnatural, so it’s best to have these takes lined up nicely at the recording stage.
If this isn’t an option, it’s possible to go through each of the doubled takes, slicing and shifting the audio to line them up with the main take. There is even software like VocaLign that allow you to automatically line up several vocal takes. However, the more aligning that can be done during recording, the better.
Anything that causes the doubled takes to stand out should be eliminated. Any time that the doubled takes play without the main layer will cause the layers to become disconnected, causing the doubling to become more distracting than the benefits it offers. Be wary of sibilant sounds especially, as a bunch of misaligned t’s, for example, will stick out like a sore thumb.
Also, keep in mind that aligning the takes too closely can potentially cause some phasing issues. If this occurs, the phasing note can be detuned by a couple cents on one of the takes to alleviate the problem.
In a music industry where a pristine, perfect vocal is the norm, it’s reasonable to be frustrated when you don’t hear the same quality in your own vocals. But like a frog in science class, all it takes is a bit of dissecting to understand what it takes to bring your vocal there.
Whether you take advantage of Nectar 3’s Vocal Assistant or make manual adjustments yourself, adhering to the characteristics that we’ve mentioned can significantly help in taking your vocals to the next level. Keep comparing with reference tracks, keep tweaking, and before long it’ll be your vocals that others will use for comparison.