What is production? Part 3: Recording
Setting a noise gate for a bass guitar with amplifier noise
A simple 8-mic drum mix, with video
How to double track easily and efficiently
Is there such a thing as Photoshopped audio?
A brief introduction to acoustic treatment
Audio problems at the BBC - TV drama audiences can't understand what the actors are saying
Recording acoustic guitar in stereo - should you use spaced or coincident mics?
How to record or amplify the melodica or any unfamiliar instrument
Do you need more plug-ins? Or more skills?
Subscribe to access our latest, up-to-the-minute articles with hints, tips and adventures in audio in the weekly Audio Masterclass Newsletter.
A question from an Audio Masterclass website visitor...
I'd like to know, apart from microphone choice, what will make the most difference or improvement to my recording?
This is a very useful question. We often find that home recording studio owners put a lot of their attention into things that don't really make all that much of a difference to their recording.
For instance, it is extremely unlikely that your choice of analog-to-digital convertor will make any difference at all to the marketability of your work, or whether a listener likes it or not.
Similarly, although your choice of microphone preamplifier will make a subtle difference to your sound it would be highly unusual for the success or failure of a project to depend on preamp choice.
When it comes to microphones there is more of a difference, but not as much as with microphone positioning, which can make far more difference than which microphone you use.
But if you want to focus your attention on the things that matter more, well firstly you should spare no effort to make sure that your song is absolutely the best song it can be. Rewrite the lyrics until they cannot possibly express your ideas any better. Fine tune the melody until it is the best tune you can write. Work on your vocal and instrumental performance until you can sing and play the hell out of your song. All that before even thinking about starting to record.
I could at this point mention the arrangement, but since this often takes shape during the recording process, I'll leave it for now.
So at this point, what is the ONE thing that will make the most difference to the success of your recording? It could well be your choice of key. If you don't choose the very best key for your song, then the end product simply won't be as good as it could have been. Let's explore in more detail...
Let's suppose that you wrote the song on guitar. That's a common expression that means that you used a guitar to help you auralize the notes and harmonies of the song. You might have chosen the key of G because it's easy to play on the guitar. The main chords in the key of G major are G, C and D - all of which are easy to play.
Many songwriters don't go any further than this. They play in G, or whatever key first came to mind, and then arrange their melody around that key. If a note is too high to sing, the writer can choose an alternative note that fits the shape of their tune.
This can work. But it might not be the very best key for the song, or the singer. A higher or a lower key might work better, and both are definitely worth trying out.
The benefit of a higher key is that it brightens everything up. For male singers it can often be the case that the song sounds best in the highest key the singer can comfortably manage, with the emphasis on 'comfortably'. Females can often sing high notes that just don't work well with rock and pop music. So for women there is a point where the key is high enough to sound bright, but not so high that the tone of the voice becomes inappropriate for the genre.
Going back to the men, for rock music it is often desirable for the singer to seem to have to 'reach' for the higher notes. So in this case the high notes don't necessarily have to be all that comfortable to sing.
A lower key can make a song more dark and moody. For the men there is a zone where the tone is deep and dark, without becoming too 'baritony', which may not be an official dictionary word yet but I'm far from the first person to use it. Singing in the style of a baritone or bass may be fine for opera but unless you're Bill Medley of the Righteous Brothers it may not work for rock or pop.
Similarly for women a lower key can make a song more dark and moody. There is also the point that a higher key might require the use of the 'head voice', which is fine when you want it, but the lower 'chest voice' register might suit the song better.
From the above, we can see that for any song there will be a 'Goldilocks zone' where the key is neither too high nor too low, but exactly right for the style and genre of the song. The best key might be as little as a semitone up or down from where you started out. Or maybe even four or five semitones - you won't know until you try.
If you are a skilled musician then it shouldn't be too much trouble to work out your new notes and chords. You only have to 'sketch them out' while you're searching for the right key. Clearly, you will have to relearn the song properly in the key you finally decide on.
An easier way for a keyboard player however is to use the transpose function either on your keyboard or in your DAW. This way you can play exactly the same notes and change the key to whatever you please. You can record this way too if you like. Doubtless some will say that this is the lazy option. Well yes it's good to develop your musicianship over time, but if you want to record your song now rather than some unspecified time in the future, then use whatever means is most practical.
For a guitarist, the easy way to change key is to use a capo. Once again some musicianly players will say that this is cheating. But actually no because a capo will let you play chords that are impossible using only your fingers. My view is to do whatever you need to get the best recording.
The limitation of the capo is that it can only change the key upwards. It is possible to tune the guitar down, and a worthwhile effort if you need to. However the tone quickly suffers and one or two semitones might be all that is practical.
There are limitations on how high or how low the key for a particular song can be. The first is how high or how low your singer can sing. The second is how high or how low they can sing with a tone of voice that is appropriate for the song.
Within those limitations however there will be a key that optimizes the use of the singer's 'tessitura'. Tessitura is simply the Italian for 'texture' and in a musical context it means the range of notes within which a singer has their best timbre. Finding the best key will be a process of aligning the notes of the song with the singer's tessitura. The best key will be where most of the notes are in the singer's best range. Bear in mind that some sections of the song will be more important than others and it may be the case that one line of the song needs to be optimized, while the rest can take their chances.
One thing that it absolutely vital to consider is that although one particular high note of the song might sound fantastic, the singer may only be able to hit that note a certain number of times before tiring.
So if you plan on recording several takes to comp together into one best performance, consider that the poor singer has to hit that high note as often as it appears in the song multiplied by however many takes you do.
Maybe that high note only appears once, as exemplified by the last note of 'With A Little Help From My Friends' the story of which you can look up on the web. In this case you can do as many takes as you like without that note. Then when you have the vocal in the bag, go for that one note and try and get it as quickly as you can.
Once again, it isn't necessarily the highest note that the singer can sing that is going to be the problem, particularly with females. It's the highest note they can sing with the tone quality that you want. Singing a high note with a soft breathy quality is hard to do, and if a singer can do it three, four or five times, that might be all they have in them for the day.
If you're lucky enough to be working with a trained singer, selection of key should be a straightforward process. They will already know what they are capable of, and will be aware of their limitations.
If you're working with an untrained singer, then no matter how good they sound there will be more uncertainty over what they can achieve and how reliably they can achieve it. In this case you will have to be a little more careful so that you don't stretch them too much.
I've come across this myself in the past and I don't think it's something that has completely gone away yet. But sometimes a singer will say that they have a favourite key. "I like to sing in E flat" for instance. Well it is quite understandable that instrumentalists have favourite keys purely because of the construction or layout of the instrument, but this absolutely does not apply to singers. The key of a song tells you nothing about what the highest note is nor what the lowest note is, nor where the most important notes center around. It wouldn't be a good idea to contradict such a singer in the session, but it wouldn't hurt to say, "OK let's try E flat first" then go up or down from there as necessary.
Although the vocal is the most important feature of any song, you have to consider the instrumentalists too. The guitarist hopefully has a capo so that shouldn't be too much trouble. But with other instruments the more sharps or flats they have to deal with, the more a player of average ability will struggle. String instrumentalists prefer sharps because they can use more open strings, many brass instrumentalists prefer flats because their instruments transpose into B flat (also clarinets) or E flat so that's one or two of the flats already dealt with.
All of the above is great in theory. But there is a troublesome issue in practice - To find the best key according to the principles outlined above, you need the singer to be available BEFORE the recording session. The reason for this is that you need to find the best key before you start recording the instrumental backing track.
So suppose you wrote a song and initially thought that G would be a good key, then when you work with the singer it turns out that A is better. If you had already recorded the backing track in G then you would have to redo it completely apart from the drums and percussion. You might think about using a pitch-changing plug-in but when you try it out you will quickly find out more about the limitations of pitch changing than you ever wanted to. Even if you used only MIDI or software instruments you will probably find that your arrangement doesn't sound quite 'right' when transposed in bulk to another key rather than being built up stage-by-stage in that key.
The best way to deal with this is to plan for two sessions. One will be to work with the singer purely on the music and in the process of doing so find the best key. This really is the ideal way to work. In additional to finding the key, you can work with the singer musically and get them to express the song the way you want them to, which will of course be a collaborative process. Ideally the singer should learn the song with your guidance. They can then take it away and rehearse more before the session.
But suppose that this isn't possible for whatever reason. One way to handle the problem is to prepare a skeleton backing track in versions in different keys. With MIDI or software instruments it's easy to do this by preparing one version, then others that are simply transposed using your DAW's transpose function. You will need to listen to recordings of the singer or hear them sing in person to get to know their range. Then if you prepare one version in the key you think is most likely to be correct, and one a tone up and one a tone down, then that should be a good compromise between what is best and what is practical.
You can either send the singer the backing tracks and let them work on the song and choose a key in advance. (You'll need to send them a version with the melody too.) This has the advantage that they can work ahead of the session but the disadvantage that you're not there to direct them and they might be singing in some way other than what you want. Or the singer can learn the song during the session. Experienced singers can learn amazingly quickly and this might be no disadvantage at all.
Of course, YOU might be the singer as well as the producer, in which case there is no problem with logistics. It would still be best however to decide on the key before the actual recording session so there is no risk of rushing this important process.
Without a doubt it is worth spending time and effort on finding the best key for your song. There may be practical problems and compromises may be necessary. But it is much more likely that the best key will be found through your diligent effort than through leaving things more or less to chance. And if the best key turns out to be C major - no sharps nor flats to bother with - then you just had a lucky break. Next time it will be E flat minor (six flats)!
P.S. Tempo is important too, but that's a whole other story for another day.
P.P.S. The most popular key according to Spotify is G major. Something about those easy chords I mentioned earlier perhaps?