The Making of a CD - FREE DOWNLOAD
Fixing a problem note with Auto-Tune
Recordings of speech by newly-starting Audio Masterclass students
Is your audio interface fast enough?
Can you hear the difference between a square wave and a sine wave?
What would happen if a spider got into your microphone?
The 10 rules of pan
Audio problems at the BBC - TV drama audiences can't understand what the actors are saying
One simple step you must take to make sure your masters sound really great
Q: Should I upgrade my Shure SM58 and use technical solutions for noise and ambience?
Subscribe to access our latest, up-to-the-minute articles with hints, tips and adventures in audio in the weekly Audio Masterclass Newsletter.
I recently wrote about the importance of finding the best key for your song. It is equally important to find the best tempo. The issues are rather different but if you don't find the very best tempo for your song, then your recording won't be as good as it could have been.
The first thing to consider, before you even think about starting to record, is the nature of the song itself. Is your song naturally a slow balled-style song or naturally a fast, uptempo song? Or perhaps it's somewhere in between. This might seem straightforward, but it's always worth considering the options.
A song that you wrote as a slow ballad might actually work better uptempo, or vice-versa. A golden illustration of this is the story that The Beatles' song Help was apparently originally written to be a slow song because of the emotion that the lyrics conveyed. Producer George Martin however, so the story goes, heard it as an uptempo song, which is the version we know and love today.
So before you even start to think about the precise tempo in beats per minute (BPM), think about the song first and decide how it should be performed stylistically.
You may have a song that you perform live and you want to record it. Since you have played it live many times, then clearly you know what tempo it should be, right?
Er, possibly not. What works well in a live setting isn't always the best solution for the recording studio. If your song is a fast, uptempo number, then you may seek to make it more exciting for the audience by playing it faster that its natural speed. If it's a slow number, you might judge the tempo according to the degree of attention the audience is paying you. Either way, if the audience doesn't seem to like it you might play it faster just to get it over and done with and on to the next song!
It is often the case that a studio recording will work better at a slower tempo than a live performance. This is probably because the recording can be perfected technically using overdubs and punch-ins where necessary, and the sound can be optimized with EQ, compression and effects more so than the live performance can.
I have to make this quick note before I move on. Some genres of music have defined tempo requirements even before you think of writing a song or a track. These are the various forms of dance music, not just EDM (electronic dance music) but also Latin and ballroom. When setting the tempo for your song or track, you need to research the requirements of the market to find out what tempi are commonly used. If you record your psytrance track at a techno tempo for example then don't expect DJs to have smiling faces when they hear it.
Let's say that you're a singer-songwriter and you can perform your newly-composed song just with voice and guitar, and you intend to fill out the arrangement with other instruments, real or virtual, when you record it in your DAW. Since your song is new, it will be a fresh experience to try it out at different tempi. You could do this anywhere, but it would be better to set up a couple of mics and record yourself, so that a) you can consider the recorded sound, not just the sound that bounces back acoustically into your ears, and b) so that you can listen to yourself on playback without having to think about the performance. Try the song faster, try it slower and home in on the best tempo.
There is sometimes the issue of whether you can play something at a certain speed. It is never a good idea to try to play something faster than your fingers want to move. It might be the case therefore that the tempo of your song is determined by your playing capabilities. But of course you can always practise. What you can't play today, you might be able to play tomorrow.
This is trickier because you're going to have to get your imagination to work harder if the music only takes shape as it is being built up in the recording process. But if you're starting with MIDI or software instruments, and there are no acoustic or electric instruments at the beginning of your recording process, then you can start recording at, say, 120 BPM, then play around with the tempo once you have a couple of tracks in place. When you can hear the drums and bass at least, then you should be able to judge the best tempo from that, with a little practice and experience. As we shall see later there will come a point, even before adding acoustic or electric instruments, or vocals, when you should regard the tempo as a decision that has been firmly made.
In this modern age of recording it is perfectly possible to change the tempo of an audio track, through 'elastic audio' or whatever it is called in your DAW of choice. I have to say though that I have found this a very variable experience. I can happily spend a day messing around with loops in Ableton Live, which will warp whatever tempo they are recorded at to the tempo of the session, and the results will sound fine.
On the other hand I recently tried to varispeed a track in Pro Tools by less than a semitone and the result was full of bleeps and buzzes. I'm not saying that Ableton Live is better than Pro Tools; I'm saying that changing the tempo of an audio track may work, or it may not work. In five or ten years' time it will work all the time and be a straightforward production technique. Right now however you can't guarantee that changing the tempo of an audio track will work out exactly the way you want it. So other than for short loops or special purposes my view is that once an audio track is recorded then its tempo is normally best regarded as fixed.
As I said earlier, this is easy. A MIDI track (or a software instrument track driven by MIDI data) can be varied in tempo any way you wish. So if you started recording at 40 BPM and at some point later, having added many layers of instruments, that you would prefer 200 BPM, then you could do it and no audio quality would be lost. This might be an extreme example, but it's perfectly possible.
So you can change the tempo of a MIDI recording at any time? Well yes, but it might not work quite as well as you hoped it would...
The problem with changing the tempo of a recording when you have already laid down several tracks is that subconsciously you will have adapted your sounds and your playing to the original tempo. This involves the instruments you have chosen, the chord inversions, the duration of notes, the release time of notes, the gaps between the notes and probably many more subtle influences that bind your musical performance to the tempo you chose earlier. There's no harm in trying a different tempo to see if it works, and it might, but it's a lot safer and a lot better to set the best tempo close to the beginning of your production process.
If you're making a dance music track, then just don't. Well not unless you really know what you're doing and what your audience wants. But for any other genre of music there is absolutely no need to stick to the same tempo all the way through. In fact, you could consider that it should be normal to vary the tempo and odd to stick to the same BPM all the way through.
It doesn't take more than a moment's thought to realize that humans can never stick absolutely to a precise tempo, but machines can do it with ease. So before the age of sequencing (pre-1985 or so) then music was played by humans and the tempo would vary naturally.
But even then there was a particular need for a constant, unchanging tempo, and that was created by the advent of the 12-inch single. This may seem like the history of the dinosaurs but it's worth noting that the conventional 7-inch single was designed to carry a single 3-minute song (per side), but for dance purposes three minutes isn't enough. So the 45 rpm 12-inch single was developed that offered a longer duration (there's a lot more history to the 12-inch single that is worth looking up if you're interested). So a song would be recorded and a 3-minute radio mix made, and also an extended version for the 12-inch. The versions of different duration were created through editing and - guess what - to edit in this way the tempo needs to be constant. Any change in tempo at an edit point is immediately obvious. To make this possible, the humans in the studio recorded to a fixed-tempo click track. It's amazing what can be done even without a computer ;-)
I digressed a little on the point of fixed tempi, but I wanted to stress that at that point in the history of music, a fixed tempo was adopted because of a genuine need. If you don't need to edit your track into different versions in this way, the tempo doesn't have to be fixed. So for instance you could speed up a little, just a couple of BPM, for the chorus and slow back down for the verse. You could slow down gradually at the end. There is an infinity of options, just one of which is to keep the same tempo all the way through.
It is worth spending some time deciding the tempo of your song for recording. Unless you're recording dance music then it is very possible that varying the tempo during the song will be better than keeping it fixed. Once you have decided on the tempo, then you will naturally adapt your sounds and playing style to that tempo as recording progresses.