Before we get started, I have to make it clear that this tutorial will work with whatever DAW you’re using. You could be mixing vocals in Pro Tools, FL Studio, Garageband, Cubase 5, Studio One or using any plugins whether Waves, UAD or stock plugins.
This is a concise guide so it doesn’t matter what you’re using and if you have problems applying the knowledge then spend some time learning your software or plugins.
As you may know I did a short post about mixing vocals in 4 easy steps and this time I want to give you something that is more detailed and help you learn how to make vocals stand out in a mix. That means this post is going to be very long so grab a cup of coffee, tea, wine or whatever you prefer and be comfortable.
I’m not going to cover anything related to recording as I’m going to make a post about that in the next coming days.
The key to getting the right vocal mix is to make sure you get it right from the source. What you need to know is that if you put garbage in, you’ll get garbage out. So make sure your vocal recording is proper. Once you’re certain that you have a good recording that is ready for mixing then you’ve won half of the battle.
Now the 1st thing I do is listen to the vocals and try to get a rough picture of the final results in my mind. Once that’s done the next thing is to put all thoughts into practice.
Tip – There’s no right or wrong way for producing great audio mixes, use the rules as a guide and always trust your instinct.
Gate Processing Tips
A gate processor mutes signals with low volume and it only modifies the RMS level of a signal. Most gates mute the signal completely while some allow you to set the compression to be just partial. The use of gates during vocal mixing is to eliminate background noise in parts where the vocalist doesn’t sing/rap.
Most controls found on a gate are similar to the ones on a normal compressor such as the threshold, attack, release and some have a range as well as a ratio parameter.
A gate with most, if not all, controls will help make this dynamic processing a lot smoother as compared to a simple gate that mutes the sound completely. For this to work well, I use an expander instead of a noise gate because it is a lot gradual and it makes it easy to get the right envelope settings (attack and release).
The attack time needs to be fast because a vocal can have parts that are percussive and the noise gate will open late and the word or 1st alphabet of a phrase wont be audible and that will reduce the clarity of the vocal. Another thing is to make sure the attack is not too fast that it ends up causing a click sound.
To get a good threshold, find the part where the vocals are really low in volume. Once you find that lowest point then push the threshold till you start to hear the gate compressing the signal then push it back up till everything fades in and out smoothly (natural). The main aim is to let the gate open when the vocals come in and close when there’s no signal.
The tricky part is getting the right threshold and release time. So if you’re having a problem with the threshold setting then play around with the release. Find a part where the vocalist sings the longest note and make sure you hear the whole note before the gate closes again. If the vocal note cuts then keep increasing the release slowly till you hear the whole note completely fades out. Make sure the settings are working well with the rest of the vocal.
If you’re using an expander and it has a ratio setting then just know that anything above the ratio of 1:10 is no longer an expander it’s now a noise gate. So play around with ratio settings below 1:10 to keep things organic and smooth.
This process can also be done manually by zooming the wav-form and cutting out all the parts where the vocalist is not singing. This can be a lot of work but do this if you don’t know how to use a gate.
Equalizing The Vocal
Once you have the backround noise cleaned out then you’re going to equalize the vocals to remove unwanted frequencies that may clash with other sounds in the mix. You can use a frequency analyzer to see which frequencies the singer is dominating.
Make sure that you equalize the vocal while the whole music is playing so that you can hear whether what you’re doing is benefiting the mix or not. Don’t solo the vocals when you’re adding any processing, the listener wont hear the vocals in solo so it doesn’t matter how they sound in solo just as long as they’re working well with the entire mix.
You must use subtractive eq first then boost after compression. You can choose not to follow this rule of thumb by the way. A parametric eq is a good choice because parametric are really transparent and allow you to create big boosts and cuts without messing the timbre.
Vocal Eq Settings
Start by cutting out anything below 60Hz, you won’t be needing that. You can push it even further if needed. You can find the fullness of the vocal around 100Hz to 250Hz and this frequancy range is good for cutting in some cases. Another part you need to cut is the muddiness area which can be found around 250Hz to 700Hz.
If you’re vocal has a honky or nasal sound then a cut around 400Hz to 1kHz will do the trick. If the vocals are harsh then use a narrow Q-Factor somewhere from 2.5KHz to around 4KHz. To add more clarity and presence then do a sweep from 4kHz to 9kHz till you find the right spot which is around 6kHz in most cases. Then boost that using a wide Q-Factor (bandwidth).
A narrow cut at around 1kHz to 1.8kHz can add some smoothness to the vocal, try it. To add some sparkle and air then a high-shelf boost at around 10kHz will do the trick. If you’re vocal is sounding too thin then a boost in the low mids can add the thickness or warmth you desire.
Keep it in mind that these eq settings are just a guideline, you’ll have to sweep around the spectrum to find the problem frequencies but at least with these tips you’ll know where to look. Use the frequency guide below if you get stuck:
Fullness (100Hz – 250Hz)
Boominess and Muddiness (250Hz – 800Hz)
Honky and Nasal (400Hz – 1.1kHz)
Presence (5kHz – 8kHz)
Sibilance (1.5kHz – 7kHz)
Clarity (5kHz – 9kHz)
Air or Breath (10kHz – 20kHz)
One last thing that’s worth mentioning is that mud is not always bad on a vocal, and it’s not always necessary to remove the low-mid range when mixing vocals because in some cases it may take away the presence or make it thin.
This is the only section I wont cover in detail because I did a blog post dedicated to vocal compression. You can find the post by clicking the link below:
Before you can start adding your favorite compressor. Ask yourself whether compression is needed on the vocals or not, is there a big difference in dynamic range (between the loud and soft parts), can you fix that manually? and so on…
Sometimes you may find that there’s a big difference between the loud and soft parts of the recording. Maybe the vocalist was moving away from the microphone during recording. You may find that some words are hard to hear or they’re mumbled.
If that’s the case then compressing something like that will just ruin it. Your best option is to do gain riding. You must manually adjust those soft parts to merely match the loud parts. You can either use volume automation but I prefer editing the waveform.
It will be easier for you to get the best possible compressor setting that will help the vocals sit well in the mix if the dynamic range is not too large. Compressors work really great if they’re not used as a fixing tool, but instead use them to polish a good vocal recording or any sound.
So how do you know when to compress?
There a lot of reasons to compress but in most cases you’ll find that your vocals are loud in some parts of the mix and sound quiet in some parts. That’s when you’ll need a compressor to even out the volume and keep it constant through out the whole song.
If you’re reading this then I assume you already know what each parameter on the compressor does and if not then take a moment to Google it but it’s really simple to understand.
The threshold determines at what level should the compressor start working and anything below the threshold wont get compressed. The attack and release time parameters are the envelope setting. The attack determines the time it takes for the compressor to kick-in and release determines the time it takes for the compressor to go at rest (zero compression).
How much compression is needed is determined by the ratio setting. That’s just the simple version of what each parameter does.
Vocal Compression Settings
I always start by finding the threshold level. The threshold settings will depend on the dynamic range of the vocals you’re working on.
To get a good threshold setting simply use a very fast attack, long release and set the ratio to unlimited then start pushing the threshold parameter till you hear a pumping effect then bring it back up slowly till you find the sweet spot.
With the same setting work on the envelope first then move on to the ratio. Using short attack and release time for a vocal makes it loud and energetic and helps it cut through the mix. Using a long envelope will make your vocals punchy without being loud. It depends on what you’re going for and the style of music you’re mixing.
If your vocals have a large dynamic range then use a small ratio of about 2:1 to 3:1 and a bigger ratio of 4:1 and above for vocal parts that have a small dynamic range. But avoid any pumping effect and make sure the vocals are as natural as possible.
Multiband Compression on Vocals
If you’re family with broadband (single-band) compressors then using the multiband wont be a problem. The parameters are mostly the same, the interface may be a bit intimidating at first glance but it wont take long to get familiar with it.
Mutliband compressors are mostly used in mastering but they can also be used for mixing vocals as well. More especially if you have a stack of vocals or just want to compress the high mid frequencies and leave the other frequencies uncompressed or maybe you just want to fix a problem frequency.
Multiband compressors are good if you’re fixing a problem like a nasal frequency for example, or to fix a percussive part that keeps jumping in the mix. I wouldn’t recommend it as a go to vocal processor because it can mess up the timbre of the vocal.
But if shaping and controlling the timbre is your goal then go for the multiband, or even if you just want to change the character of the vocal. A multiband compressor can be a great tool to make a “dynamic controlled” boost in the high frequencies for instance.
It’s a good tool for mixing vocals but use it with caution.
Guide For Using The De-Esser
Once you have the vocal compression done then you’ll notice there will be some sibilance on the vocals. Even a good recording will have sibilance sounds especially if it’s a female vocalist. Compression can also cause sibilance as well.
These are sounds with “sss” or “ts” and these are caused by words with alphabets like t, k, s and z. They’re are commonly known as hissing sounds.
These are not generally bad for the mix, but in some cases they can be annoying and can sound pretty obvious after adding effects such as delay. A De-esser can also ruin the clarity of the vocal so you want to add it but not remove the hissing sound completely. You need a little bit of sibilance to keep the vocals natural sounding.
A de-esser is also a dynamic processor so you’ll need to add it right after the eq and compressor but before you add any time based effects like reverb. This is just to ensure that the de-esser is only working on the sibilance and not messing with other frequencies.
It’s not really hard to find the sibilance, in most cases a female singer tends to be sibilant in the 5kHz to around 8kHz range and a male vocal tends to be low around 3kHz to 6kHz. You’ll have to search for the sibilant though, there’s no one size fits all setting.
Reverberating The Vox
This is another section which I’ve covered on the blog before. So I wont go into too much detail about adding reverb on vocal. You can check out the blog post about this section on the link below:
A common mistake people make with reverb is to focus on how it sounds, instead you should focus on how it makes you feel. Reverb doesn’t only add depth or soften the vocals but it also adds emotion.
Choosing a good reverb sound for your vocal is very crucial. Just like a multiband compressor can mess up with the timbre, the same thing will happen if you choose a wrong reverb sound that’s not suitable for the vocalist you’re working on. Try different reverbs till you find one that doesn’t change the character of the sound.
There’s different kinds of reverbs. A room reverb is really short and it will add little depth and space to the vocal. A hall reverb tends to be long, it sounds full and have more reflection.
A plate reverb carries a lot of early reflections and it has a thick sound but for a short period of time as compared to a hall reverb.
Reverb effects are really easy to get familiar with unlike dynamic processors. But the one most important thing is to find the right reverb time settings that will match the tempo of the song. You can do this by using your ear or using a simple formula.
Take 60,000 and divide (÷) it by the tempo of the song. For instance, if you’re mixing a song that’s playing with a tempo of 90BPM then 60,000 divide by 90 = 666.7 milliseconds (0.6 seconds).
One trick I learned in music production school that I don’t see a lot of engineers do is automating the reverb throughout the whole arrangement. For instance, the chorus part will be more reverberated as compared to the verses and other parts.
This can sounds really good especially with delay effects but don’t exaggerate this, it doesn’t have to sound obvious to the listener so use it with caution.
Another great trick, especially if you don’t want the vocals to sound like it’s in a room then remove all the early reflections and only use the tail of the reverb. That way the vocal will sound as if it’s dry and it will be upfront in the mix but the tail will be reverberated.
Don’t forget to use the pre-delay time to determine when does the tail start getting reverberated. This will keep the vocals present even though the attack of the vocal will sound dry the tail will have the reverb.
Delay Effect on Vocals
A plugin delay effect basically records the incoming data, which is the vocal in our case, then store it in a buffer. While old school units used to rely on tape or digital sampling technology.
Delay effects can be used in a simple form or complex patches which involve adding effects such as distortion, auto-filter or even an eq to change the character of the delayed signal or just to clean things up. Delay effects can be a great way to make a doubling effect on vocals. Most Rock engineers use the delay effect instead of the reverb to help the vocals to sit well in a mix without pushing it back or making it sound distant.
Most delay effects will have a way to set the timing of the delay effect, which is very crucial. Some will come with a sync tempo button while in some cases you’ll have to use your ear or simply take 60,000 and divide it by the tempo of the song.
The aim of using a delay effect on vocals is not for it to be heard but just enough to support the vocal and make it sound bigger and never make the delayed signal louder than the original signal.
Short delay times of about 80ms work well to blend the vocals with the entire mix, especially if the reverb is making your vocals too thick and ruining the clarity then you can use a short delay instead. Be careful with using long delay feedbacks they’ll add some muddiness.
If you’re using the delay effect on a send channel then keep it at 100% wet and if you’re using it as an insert then 30% is a good starting point then play around with it till you find a good spot.
The most commonly used delay is the Ping Pong which is alternating echoes that are panned hard left and right in the stereo image. To achieve this effect you need to make the delay time on the left half of the right side’s delay time. Then the vocal will bounce around the stereo field, from the center (original sound), to the left then to the right channel.
If a stereo delay effect is adding mud on the vocals then go for a mono delay. Alternatively you can use a reverb with a long pre-delay time of over 120ms.
Another great sounding delay trick is the ducking delay effect. To achieve this effect you’ll have to add a compressor on the delay channel then set the side-chain input to be the vocal aux send. Use a fast attack and slow release time then set the threshold and ratio to taste.
Then the compressor will close the delay when the vocals are playing and open as soon as the vocal compression goes to rest, then the last phrase of that vocal part will echo.
Finally, if you’re using long delay times then don’t add it throughout the whole song, use it in different parts of the mix or in certain phrases.
Quick Panning Tips For Vocals
This section is going to be the shortest because it depends on the material you’re working on. But basically you want to keep your lead vocals at the center especially for the verse.
The chorus part needs to be wide that is why it’s recommended to record many takes and pan them left and right. I usually keep everything below 25% pan, I never go to 100%. Panning one vocal part hard left and the other hard right (100%) is like having a single mono channel except that it will be 3dB louder.
Back in the days it was good to pan things hard, especially for background vocals to make them sound like they’re at the edge of the speaker. People are using ear buds, so making your vocals too wide might not sound good on these ear buds. Go for something a little tighter not wide, anything less than 25% pan will do.
Pan the chorus stacked vocals according to taste but make sure non of your vocals disappear when the music is played in mono. If the stacked vocals were recorded like a choir where you have the brass, tenor, soprano and bass section then keep the voices will a low tone in the center and high harmony vocals on the sides and if there’s Adlibs keep them in the center.
If all the stacked vocals sound the same then pan them according to what you feel sounds good. Things you can pan are mostly the harmony vocals and backing vocals. Experiment with panning and make sure to test your mix in mono.
Mixing Background Vocals
Just like the word says – “Backing Vocals” that means they need to be at the back. Your effects and dynamic processor settings for the backings will differ from the lead vocal settings.
For instance the compressor will have a fast attack with a medium to long release because if the release is short then the backing vocals will be energetic and loud. That’s not what you want for the backings and that’s why you must use a long release and fast attack to keep them punchy but never loud.
Don’t allow you’re backing vox to have too much dynamics, use drastic threshold and ratio settings but avoid any pumping effect. You’ll also need to use a lot of reverb to push them at the back of the mix and be careful not to drown them with reverb.
Long reverb and delay times work well because backings don’t play throughout the whole song. Use a stereo image effect to widen-up the backing vocals in the mix, this will make them less direct helping the lead vocal to lead.
Don’t make the mistake of over-processing the backing vox, remember that they have to support the lead vocal so they don’t have to be pushed far back, they also need to be audible and clear.
For the eq settings I normally use a drastic high pass filter at around 150kHz and a big dip in the low-mids. Sometimes a dip around 1kHz does open up some space for the lead then I’ll add some sparkle using a high-shelf. This high-shelf will be a bit more than the lead though.
Push It To The Limit
A limiter is another great tool you can use for compression. Unlike a compressor which acts on a sound as it arrives, a limiter has a look ahead feature with tends to make the compression a lot smoother even if it’s pushed hard.
A limiter usually uses a fast attack, mostly a hard knee and unlimited ratio. It’s a special type of compressor that will never allow the input signal to exceed the threshold.
Sometimes using a compressor can squash the vocal too quickly and cause it to distort and in that case the limiter is a good option as it has a look ahead feature which allows the limiter to see a few milliseconds before the compression takes place.
Usually a limiter should be the last thing you add in your vocal chain after all the processing is done. Use it to trim out the loudest peaks or if you want to push your vocals to the front of the mix.
Sometimes you might find yourself using heavy compression with lots of gain reduction, a fast attack, high ratio and a lot of make-up gain. In that case, technically you might be using the compressor as a limiter. So try a limiter instead and see how it sounds. You might find that the limiter is working better because you were using the compressor for what it’s not designed to do (if that makes sense).
Using a limiter is much faster and easier especially if you’re not familiar with compressors. The vocals will be right in your face without fluctuating.
Use a limiter to help the vocal sit on top if it’s getting lost in some parts of the mix. It’s little things like this that make a big difference in the mastering stage as non of the peaks will keep jumping in the mix. Which will make it a lot easier for the mastering engineer to polish the mix.
Remember, less is more so don’t squash the vocal.
Producing great sounding vocal mixes is all about experimenting. Things like tuning your vocals can also help the vocal blend well with the entire mix. I would advise you to get the timing and correct vocal pitch from the source instead of relying on tools.
There are a lot of great tools out there for fixing vocal pitch and timing issues but the most popular ones are Melodyne, Waves Tune, Cubase VariAudio and AutoTune. Another great tool that can add some sparkle to your vocal mix is the Exciter. I don’t use an exciter on the lead I usually use it on the backing vocals.
Also use modulation tools like chorus to double up sounds. Effects such as phaser and flanger can add drama to a song when used in places such as the breakdown of a mix. Modulation plugins work really well on background vocals as well.
What comes 1st between compressor and eq really depends on the vocal recording. If you’re going to cut then use the eq 1st and if you’re going to boost then use the compressor 1st. You can use 2 equalizers, one to cut and one after the compressor to boost.
Parallel compression is another great technique you can use for producing good dynamics for your vocal mix.
I think that’s it 🙂
For once I thought I should write a more detailed blog post instead of the short ones. Hope you found it useful and if you guys prefer short posts then let me know. Long posts like this one tend to take long to write but if that’s what you guys want then I really don’t mind at all.
If you would like to add something I might have missed then do so by leaving a comment below.
Mixing Vocals Using Waves Plugins – Full Webinar with Yoad Nevo
Happy Vocal Mixing!!!