So what makes a vocal sound so good that it leaves you speechless? Well other than the performance which is up to the singer…. it’s the tone and quality of the vocal. What makes the tone and quality good is a well balanced vocal. The right amount of lows, low-mids, mids, high-mids, and highs.
So how do you achieve this? ….I do it with my ears. After lots of practice, your ears can pinpoint if something sounds off balance. Listen to the vocal with and without the music to determine what needs done. You should be able to find things you don’t like. Maybe it’s not bright enough, maybe it’s too bright. It could be muddy and muffled, or it could need more warmth.
My tip to you is to do one thing at a time and don’t do things too drastically. Handle ONE problem at a time. Look at it as “building the vocal sound” by carving it and shaping it to accommodate your ears and what they want to hear. If you are unsure what frequencies need work to get the vocal balanced, check out my ebook called The Ultimate Vocal Production Handbook!
So this is sort of a debate between audio engineers. When you ask people how long they take to mix a song, I have heard a variety of answers. Some say an 8 hour day. Others have said two 8 hour days, some say half a day. So what is technically correct or just a good approach in general?
In my early days I did a lot of rock and metal so an 8 hour day sometimes was the case because of drum replacement, drum editing, pitch correction, cool post production effects etc. However nowadays i get more hiphop, pop, and r&b so its basically just mixing instrumental stems and vocal layers.
So after years of doing this, I have found that less time on a mix results in a better mix.I have done a mix in 1 hour and loved it. Sometimes 3 hours and loved it. Sounds crazy but what your ears naturally hear off the rip and the decisions you make are often the best decisions. Those times where I obsess over the mix and let high expectations of the artist pressure me, are the mixes that don’t turn out right. Mixing is all about trusting your judgement and not thinking too much about what the artist or label will think. You obviously need to have a reference and a goal of how the mix should turn out but don’t over think.
Ask yourself if it sounds good to you. If you were the artist, would you be satisfied? That goes a LONG way.
This article is targeted towards the non-Pro tools users! I have noticed that programs other than Pro tools have a bad visual for editing audio. Logic, garageband, presonus one, etc are just a few that come to mind. It is very important that we can visually see the gridlines and that we correctly set the tempo right.
First you MUST find the correct tempo of the song and make sure all of the drums and instruments are quantized to perfect time or near perfect time. If this step is done, you should see the gridlines in Pro tools lining up with drum hits etc. Although there may be a way that I’m not aware of, but when I have used Logic before, I could not really see where the grid lines were to be able to edit audio.
If you can’t see where the downbeat is located at, it will be hard to copy paste chorus vocals, or will be tricky to chop and move around vocals that are off time. I have always been a Pro Tools guy because the editing window is so easy on the eyes! Although Logic and other programs may look easier to use when it comes to other things like a mix window or general functions, the edit window I have found almost useless.
Today I want to write about something that I had to learn over the years. One of those things where when you know the rules….you can break them! I always thought that using 1 de-esser plug-in was some sort of rule. If I used more than one then it would hurt the audio. While that can be true if you don’t know how to go about it…it also can keep your hands tied if you are struggling controlling the sibilance and trying to just use one de-esser!
No one ever taught me this, it was something I had to try for myself and experiment. So what I like to do is set one de-esser in the high-mids at around 4k. I select wide band setting because it sounds more natural to me and also compresses frequencies that are nearby as well. Bring down the threshold until you hear the sibilance tighten up just a little. If you are crushing it and taking away the sibilance completely, it’s too much.
If the one plug-in doesn’t do the trick, load a second de-esser. It can be the same plug-in or a different de-esser plug-in. Set the frequency to a higher frequency further away like 6 or 7k. Bring down the threshold until you hear those higher frequencies in the sibilance get smoother. You can try wide band to start, and resort to split, or narrow band on the de-esser. Whichever sounds the best is the keeper. The goal here is to have a natural sounding vocal. If you want to try a third de-esser in another frequency area, do it but be subtle. Just control the ess’s…that’s it!
As an engineer, you should do your experimenting and playing with plug-ins in your off time or when mixing the song. During recording sessions, it’s best to put on some plug ins you know that will work for the time being and move forward. It can be a real vibe killer for an artist that is amped up and ready to record and you are trying out different plug-ins, effects etc. It’s one thing to adjust the mic or maybe swap out a mic if it isn’t working for the singer’s voice, but everything that is post processing like plug-ins, can wait.
One thing you can do is save a vocal recording plug-in template that you can use for every session. Then it’s very simple. Just import those settings in 30 seconds and you are ready to record with a nice sounding vocal plug-in chain in tact. In Pro Tools you would do this by creating a new session, importing some vocals, mixing the vocals with plug ins like eq,compression, and reverb, and then saving the session’s name as “vocal recording template” or something. Then when you go to record you next client, just hit Shift + Option + I. This will allow you to import the session data from your template. Unselect import audio so you only import the channels with plug-ins.
So here is something that took years to finally figure out because no one teaches this….tuning the doubled vocal less than the main vocal. So obviously you want the main vocal to be pitched very tightly so that the listener isn’t turned off by a sour note here or there. This is a good mindset to have, you should always pitch very tightly while still sounding natural. However with doubled vocals it can be good and sound better to tune them not as perfect. Don’t get me wrong, it still has to be very close to “on pitch” but not pitched as tight as the main because the two vocals will sort of phase out because they are too similar. Also it won’t really give the doubling effect as much and becomes pretty pointless.
A good method to achieve this is to draw with a higher default retune speed setting. This method only works if you are drawing with autotune in graphic mode. When drawing the double vocal’s pitch, go to options and set the default retune speed for the object tool to something higher than what you did for the main vocal. So if you had a default retune speed of 25 when doing the main, try 50 when drawing the doubled vocal.
When using Melodyne or Waves tune you just have to use your ears and move the blobs around without straightening the pitch too much with the pitch modulation tool. Overall it is way easier to not tune the double too tight with these programs because naturally these programs are designed to not tune the vocal very tight leaving it more natural in general. You could always draw the main vocal with Autotune and draw the double with Melodyne. I do this all the time and get superb results!
One thing that I have learned over the years is that you can’t hear that well in headphones while recording. You also don’t want to turn them up too loud because then you will get headphone bleed into the mic.
There’s 2 methods that work wonders though. The first method is the “one ear off” method. This is very common for singers to do while recording. It works very well. Take one headphone off, press that headphone against your head to reduce headphone bleed into the mic, and you will hear your voice well! You will also hear your self with no latency. Real-time monitoring will assure that you are singing the parts to your full advantage! This is even a common technique for singers at a live performance. Many singers will take one in-ear monitor out to hear themselves better.
The other method I have found super helpful is what I call the “press firmly” method. With this method you simply take your hands and press the headphones firmly against both ears. This will increase the volume of the headphones, bring in more low end to make your voice sound fuller, and reduce headphone bleed into the mic even more! If you are the type that doesn’t need one ear off, I would recommend using this method at all times! As long as there aren’t a lot of plug-ins causing latency and the buffer settings are low, there should be little to no latency!
This is a huge debate with vocal instructors. Some think you are limited to a certain vocal range that you were born with and others believe you can have a 5 octave range! I follow Brett Manning and he believes we are able to have that 5 octave range… male or female…it’s possible!
He is right because the number one thing that limits singers from having a big range is not being able to get out of their chest voice! I get it, it’s very natural to sing from your chest because that is where you talk from. All day everyday you are use to talking from your chest. This can definitely make things tricky for you. In order to get out of your chest and into your mixed voice, you need to practice resonance.
What is resonance? Let me explain…
When you hum a melody, your face should vibrate and almost give you a tickling sensation because of the vibrations. If you don’t feel the tickle, then you must practice humming. Many vocal exercises have you humming scales for a reason! If you can get good at humming or in other words….vibrating your face, then you can create great resonance in your singing voice. Resonance will allow you to hit those super low notes and those very high notes.
There is another factor though that some people still struggle with…. and that is vocal chords. When singing high notes you need to slightly compress your vocal chords to avoid strain. On low notes you need to relax them more how you would when you are talking. Don’t let this last part throw you off, practice resonance/humming and you will be surprised on how your range improves!
Have you ever pondered about why your mixes don’t sound quite as loud as other songs that you hear? That’s what is called professional mastering. The song was likely mastered by someone that understands frequency balance, how to control certain problem frequency areas, and knows how to get loudness without over-compressing.
I know what you’re thinking….you may not be a mastering engineer or even close to that level yet, but there’s another way to get it close if you are on a budget and can’t afford mastering. First off, with this method you want to make sure your mix is very very good. That’s the catch! It should be a very balanced mix that sounds good at low volumes and high volumes.
Do some light eq on your mastering chain. Roll off 20 hz and down, put a 1db boost somewhere in the sub lows, boost 0.5db somewhere in the low mids, and put a high shelf at 12 or 13k and boost about 1db. These are subtle moves that will make a pretty good difference. Now that you have done that, if you have a multiband compressor, place that next and select a mastering preset. Bypass the plug-in on and off to see if it tightened the frequencies up.
Now you are ready to grab 2 limiter plug ins. They can be the same limiter plug ins or 2 different. I use the Waves L2 first and the Waves L3 ultramaximizer second! Use your first limiter to do the heavy lifting. I normally can bring the threshold down to -6 without getting any gain reduction. The point of the first limiter is to make things louder without gain reduction. Second, grab your second limiter and pull the threshold down until you get 3 or 4db of gain reduction. If the mix sounds okay to push it a little harder then try it! At this point you have a very loud master. Make sure dither is “off” on the first limiter and “on” for the second limiter!
I was mixing a song today for a hiphop artist and I did a technique that I have to do from time to time. The instrumental had some really driving synths and tons of high mid frequency energy throughout the track. When instrumentals don’t leave room for the vocal it can be a challenge to mix the vocal in.
There are a few ways to go about this. First of all, if you have the beat stems…which I did not, you can easily turn down the element that’s getting in the way or eq that certain element to make room for the vocal. Most of the time when working with hiphop I don’t have the stems and I am forced to work with the stereo track. That was the case for the song I mixed today.
So what you can do in this scenario is put an eq on the instrumental track, use a narrow Q setting and try cutting out some high mids. I like to start around 3 or 4k and cut by 3 db or so. Make sure its not taking away from the attack on the snare too much. Normally the snare can afford a cut in the high mids and will still be okay. If that cut still isn’t leaving room for the vocal, then try another cut either above 4k around 5 or 6k or below 3k around 1 or 2k. If you want the vocal more pronounced with lyrical content, go with 1 to 2k. For a brighter vocal cut a few db of 5 or 6 k in the instrumental.
Another method I use from time to time is a plug in called Trackspacer. This plug in will duck down the frequency in the instrumental every time the vocal is going. To set it up, insert Trackspacer on the instrumental channel,choose a bus in the Trackspacer plug in, then use an aux send from the vocal channel to send signal to that same bus. Now every time the vocal is going, the instrumental will duck out a bit. When the vocal isn’t going, the instrumental is untouched.