By Sean Shannon (MixMonsterz.com)
Sean is a touring/studio drummer, mixer/producer/engineer, songwriter, and educator based in the Orlando, Florida area. Sean has made music with artists like Molly Hatchet, Billy Payne, Pat Travers, Brian Howe, Les Dudek, Rob Rock, Impellitteri, Charlie Daniels, and many others, and has worked with producers like Dennis MacKay (David Bowie, Judas Priest), Malcolm Springer (Collective Soul, Faith Hill), and Steve Thompson (Guns n Roses, Metallica). Sean attended SUNY Fredonia’s Tonmeister Program, studying audio production with Dave Moulton and percussion with Dr. Theodore Frazeur before moving to Florida. Sean founded The Mixing Workshop, an audio ear training and mixing course for audio engineers, and writes articles about mixing for Recording Magazine.
(Be sure to post and let Sean know if you have any questions or if you enjoyed his article)
Tips For Mixing Vocals
The vocal is arguably the most important element of a mix. The ear focuses on the vocals, and the girls listen to the lyrics. Once the vocals are recorded and ready, how can you maximize their emotional impact?
Let’s face it. Getting the vocal to sit just right in a mix is not easy. Sometimes it seems like the vocal track sounds so good, all you’ll have to do is pull up the fader, and it should sound right. But many times (after getting the music track slammin’), “shoehorning” the vocal in the mix takes some creative engineering.
Frequencies pop out, “ouches” may hit you in the midrange when you turn the mix up, “sss’s” may shear your hair off, and “P’s” may be a gut punch. Words and syllables may seem to jump out or get buried.
What happened? The track was recorded well! Nothing happened, it’s just that the voice is the most dynamic instrument, with a wide range of energy at various frequencies, and it benefits from some professional treatment.
While every style of music may differ in approach to treatment with regard to level and effects, most have one requirement: the listener must be able to hear and understand the words. Make the lyrics clear, and deliver a sound that brings the listener into the story, and you have a winner.
Here are some tips for working with vocals in a mix. Using the standard tools available in almost all DAWs and studios, we’ll go through the process of getting the vocals balanced and sitting right in the mix.
Learn the words
That’s what a fan does. It’s important to learn the words of the song, have a complete lyric sheet handy, and make notes on words or phrases that need attention. By complete, I mean every word is written out instead of ‘repeat chorus’. This allows the engineer to make notes on any word in sync with the timeline of the song.
Once the issues have been noted on the lyric sheet, it’s easy to target them for treatment. Typical tools include automation, either in the DAW or on a console, compression and eq, and effects.
Back in the day, the whole band would belly up to the console and grab a set of faders or sends and make their moves on the mark as the print goes down. The marks would be on masking tape by the faders or send knobs, and each mix was a performance by everybody, all hands on deck.
Nowadays, the computer becomes those extra hands, each expertly performing every move to exact precision. This is called automation, and just about every aspect of your mix can be automated. You can automate levels, eq, effects, just about everything to make the track sound perfect. Like a bandmate, you just have to tell it what to do, step by step, using automation.
If, for example, a few lines seem louder than the rest, you can automate the volumes so they all match. If there are certain words you want more effects on, you simply raise the effect send on just those words. It’s also easy to automate eq settings to turn on and off on problem areas, too. Automating the eq can be a real mix saver, just specify the parameters you want to automate, like ‘bypass’, and write in the changes in the track’s automation lane, in sync with the timeline.
The most obvious use of automation is for riding fader levels. (See fig 1 for an example of volume automation helping ‘plosives’, or low frequency booms from the singer’s air hitting the mic) Some words and syllables need help in a big mix, where plosives are held back and whispers and tails are up front for impact.
If you are using an analog mixing board, you can also mix in sections, with live fader moves (mixing by hand), by printing back into the DAW on a stereo track and punching in for each section. I use a large format analog console, and I combine the onboard mute automation of the console with the automation in the DAW for a quiet mix, and I ride faders as the print goes down if necessary.
Another great trick is to split the vocal up onto a couple of tracks, so you can eq, compress, and effect the chorus vocals separately from the verse vocals. This can be done by duplicating the vocal tracks on the screen, or by sending to multiple channel faders on the console, then mute the parts you don’t want for each section of the song. I typically use a Verse track, a Chorus track, and an Effected track for special effects vocals.
Compress for Success
Using compression on a vocal can really add size and stability to the track. Used skillfully, compression can bring out the articulation, the details in the words, while holding the singer firmly in place on the soundstage.
If I could only use one compressor on a vocal, I might use a super fast attack and release time to keep things from jumping out while getting out of the way quickly. The Distressor is very good at this. Set the threshold so the vocal is almost always in compression, with peaks getting a good 6 or8 db of gain reduction, then raise it from there if you don’t like the sound.
I look at compression as either fast or sloppy slow. Slower attack and release times keep the average levels controlled, but peaks jump through. One tip is to use two compressors in series, chained one after the other. Use one to gently control average volume with a slower attack and release, and another to bang back the peaks with a fast attack and fast release. Keep gain reduction reasonable on each one, and tweak between the two of them until you get a clear, full sound and a consistent vocal level. This should help make the vocal sit in place better. (See figure 2 for compression settings)
Parallel compression can be a useful tool for lead vocals. Send the vocal to a compressor, and mix that squashed signal back in with the original vocal on another channel. This can increase the apparent size and clarity of the voice.
Some engineers prefer little to no compression, preferring to automate or physically ride the faders. Others rely on compression heavily for their sound and control of the vocal. Find the method that works for you and for the song by learning the sound of compression and comparing your mixes to pro mixes, and analyzing the vocal treatment.
De-essers are also typically compressors, but are limited to acting upon a single band. You can listen to the frequency band being affected, and search for the most annoying frequency. Set the amount of reduction so that the undesired frequency is brought under control, but the vocalist doesn’t develop a lisp!
Many times the de-esser can dull the vocal while killing the harshness. Put the de-esser before the eq, and carefully boost the top end (12KHz and up) of the vocal with a shelf eq.
A multiband compressor can be useful on a vocal in the respect that, like a de-esser, it can act on one band, or more than one band, separately and simultaneously. With vocals, the energy in the various frequency bands changes dynamically throughout the song, so a multiband compressor can hit low mid and other high frequency jumps selectively, and only when they occur.
Equalize for Size
A careful use of eq can really put the vocal in its place and help control dynamics before any compression is used. For an easy example, rolling off the lows before the compressor can help with breath pops that might cause overcompression.
The wide range of dynamics and frequencies in the human voice almost guarantees that eq will be needed at some point throughout the song. There may be notes that stick out too much in a certain frequency range, or areas of the track that are too dull or boomy against the music mix.
Many times, when a mix is turned up, the vocal will seem like it’s at the right level, but is a bit harsh, or hits your ear in the mids and upper mids. Eq can reduce the harshness while maintaining the clarity. (See fig 3 for a vocal eq setting)
Common techniques for eq’ing vocals for clarity include rolling off the lower octave or so (under 80Hz) and finding and gently cutting with a medium bandwidth in the 300-600Hz range. Find an area in the 900Hz-1.8KHz range with a med bandwidth, and give it a slight boost. If necessary, roll off the highs if the mid boost accentuates the consonants too much, find a balance that leaves the vocal clear but not harsh. Sometimes spikes still occur in the sensitive area, the upper mids.
If a de-esser can’t tame those spikes, then sweep through the upper mids on the eq with the cut fully engaged. You’ll find a spot that will still allow the vocal to sound somewhat natural, even with full cut, but without the harsh spikes. That is the area to be worked on. Without changing the frequency, dial some gain back in, and narrow the bandwidth of the cut until the vocal sounds full and natural again.
This may make it seem slightly dull at first, but will allow you to turn the vocal up a bit without seeming too loud. You may want to add some highs at 15KHz or above to open the sound back up. I use multiple eqs at the beginning, middle, and end of the chain, each performing a function relating the next processor in line.
One important note for mixing vocals is that the music track must be able to accommodate the vocal. In other words, keep an eye on the frequency content and individual track levels of the music mix, making sure to leave room for the vocal in the critical midrange. This is called “spectral balancing”.
For example, if the guitars are too bright, or have too much energy in the 1-5KHz range against the vocal, they will require that the vocal be pushed too loud and risk becoming too harsh, just to be heard clearly. Keep the mid frequencies under control on the music mix, and the vocal will sit in the mix better.
If eq’ing the vocal inflicts too much damage overall to fix to occasional problem frequency, it’s easy to automate the eq so it comes on and off on certain words or syllables, or cuts a frequency at certain lines in the song.
A neat trick is to duplicate the vocal track, using one channel for verses and another for choruses, so each track can have it’s own eq, effects, and compression chain. Just mute the sections you don’t want to hear on each track.
Use effects for impact, size, and to blend the singer with the band. Ambience, reverb, and delays can work wonders on a vocal, as can a little chorus. The contrast between a dry (no effects) vocal in your face, and a lush vocal with a full space around it and some delays can be used to your advantage during a mix.
Each section of the song can have differing treatments. The a’ cappella breakdown might be dry, while the full band chorus may be better delivered with the vocal wet, with delays timed to the song and a bit of reverb to put the singer back in the band.
A common vocal effect is to use a pitch/delay effect, like the Dual Shift in the Eventide H3000, in stereo on a vocal. The effect makes micro-changes to pitch and timing in real time, adding a nice, thick fullness to the vocal. There are plugins that do similar things for you.
Send the lead vocal to the effect on a send, and bring up the stereo return of the effect until it widens the vocal. It can really make a vocal come alive, but be careful, it’s easy to overdo it!
Another cool effect for a vocal is distortion. Adding a little bit of distortion, like an amp sim or tube plugin, and blending it in with the original vocal can create some very compelling vocals in a heavy track. You could automate the effect so that it turns on for certain sections of the song, or gets heavier in others. Telephone voice is another useful effect. (See figure 4 for this effect)
In addition to the tools described for lead vocals, the backup vocals may require additional processing to fit in. Usually, more lows can be cut, and consonants and sibilance can be attacked more aggressively, since the listener will get all that from the lead vocal. If the consonants, like “t” at the end of words are off time, they will be exaggerated, and harsh frequencies and sibilance adds up.
Tame highs and lows with shelving eq, and automate the volumes of the words to support the lead vocal and reduce clash with consonants. A compressor on each vocal track helps blend them, then you can put a stereo compressor across the backup vocal group to smooth out the blend even further. Use your ears, and set the compressors for control with clarity, using the attack time to control how hard the first consonant hits, and the release time to smooth sustained notes.
Pan the backup vocals in a stereo spread around the lead vocal, and put some effects on them to give them their own color. They can get away with a bit more reverb (or ambience) than the lead vocal, putting them behind it. The lead note should be the dominant vocal in the blend.
Monitor for the real world
Try to set vocal levels, rides, and effects levels while listening on small speakers. That makes it easy to hear if it’s too loud or too soft. I also engage the mono button on my console while setting balances. Making the vocal balanced in the mono sum will guarantee clarity over the internet or phone speaker, and will usually work very well once you turn it up in stereo.
Reduce your monitoring environment to the lowest common denominator, and make sure the vocals all work on that. It’s also a good way to check levels of competing instruments against the vocal.
Here’s a checklist that sums up what we have covered in this segment on mixing:
TIPS FOR MIXING VOCALS:
1. Have a lyric sheet handy for marking phrases and issues to be addressed.
2. Use automation to control all aspects of the vocal, including level and eq/effects changes throughout the song.
3. Use compression for holding the vocal in place, and for fullness and clarity.
4. Use eq as a tool to remove unwanted energy from the vocal while preserving the intimacy and fullness, and to make room in the music track for the vocal.
5. Use reverbs and delays as needed for space and dynamic contrast or to help the vocal appear to sit onstage with the band.
6. Treat backing vocals as support, blending rather than sharpening.
7. Set vocal and effect levels on tiny speakers while monitoring at low volume in mono so the vocal sound good everywhere.
Portions of the content contained in this article appeared in Recording Magazine , and are used with permission from the author. Contact Sean at www.mixmonsterz.com.