Using vocals inside your house music productions

Here's probably one of the hardest topics to cover when it comes to making music in general. At least, it is hard for me.. let me put it this way. Using vocals inside your music productions is one of the hardest things to do, because they can go out of sync very fast. There are 2 common problems with vocals: timing and pitching the sample right. But since pitch also controls the runtime of a single sample, the 2 problems are related in some way. The same problem will also happen when creating a hip hop loop (based on a single wav file). It sounds nice in pitch, but when you want to make it fast, you have to pitch it up again and then the sample goes faster, yet the sound does not sound at the right tone height.

The problem also works the other way around. For instance if you have a vocal timed right on any beat, it sounds nice and goes along with it. Then you make a melody or bassline and the melody and the bassline play along with the beats, but for some reason the vocal starts to sing along as it wouldn't be a part of the song itself.

Personally when I listen to most of the house music pieces that are made by amatures, I feel like 5% of all the 'wannabe' music artists understand the problems I discribe in the just mentioned section, however I feel like 95% of them are not even aware of the problems. They basicly think to themselves: "this sample just sounds cool" and drag it in there. Yeah, the sample sounds cool, but it makes the rest suck...

So, the first question you might want to ask yourself is this: do I really need a vocal sample inside my music production? The answer would be no! It is not a must to have vocals inside your house music production. There are tons of songs that don't have it. so if you are trying to place a vocal inside your production, and you come with the problems I just described, you might want to think twice to place it there. If it does not work out, the better way is to get rid of it to begin with.

Vocals are usually a pain in the butt

For those who really want to take time to place their vocals inside your house music production, well... here we go. Welcome to musical hell ;)

Before I am getting in the whole topic itself, let me also add: Propellerhead Reason does not allow you to record any vocals using a mic. To solve that you could use all kinds of different tools to record it, and save it as a Wav or Aiff file and import it inside reason. Since I have had this question for so many times I also mentioned this inside my blog entry called Frequent Asked Questions

As I have said in the prelude of this article, there are 2 problems we will be dealing with, timing and pitching. For the timing issue we can use 2 different tools. First one would be using recycle and slice the vocal up into small pieces. The saved vocal will become a rex file and we can import that rex file inside the Dr Rex or NN-XT sampler.

When you want to use recycle as a vocal import module, one big tip I want to point out: slice it heavy! The more slices you have, the easier it becomes to create a tempo changes with it (since the slices are played faster after eachother and we don't really have to go heavy on the pitch changes). Be aware though, that a Dr Rex player can't import rex files that have more then 92 slices. As the orignal vocal I am using inside the next example is recorded on 121 BPM I will increase the tempo to 132 BPM to make a simple point with it. as you will see in the next example file, the speech goes fast, but it does not really have a 'big' problem with it. the only real big problems are probably pattern 5 and patter 7 in the next sequence:

Download RNS example file: Vocals with recycle
Download MP3 example file: Vocals with recycle

The real nice part with slicing up a piece like this is that you won't really have many problems when the tempo changes from 110 beats per minute till 135 beats per minute in this case. Since it still feels and sounds like it is a fluid rhythm that is going on. This is also one of the most common setups I use when it comes to vocals. The reverb (in this case the RV7000, but can also be an RV7) will reduce the gaps in parts where the vocal has a stoppage time. The delay will be a rhythmic mechanism, so in fact when the beat would drop it still feels like the beat is still there, even while it isn't. To have this rhythmic effect one could use delays on 2 steps, 4 steps or 8 steps. 16 steps might also work, but it can feel like the delay is a bit 'far' away. for short words like "Lets Go".. a 16 step might work in this case, but having a long vocal as I have presented in this example, it might become a bit too messed up.

Download RNS example file: Vocals Lets go
Download MP3 example file: Vocals Lets go

In the lets go example I am infact using a 16 step delay time. Which works rather well in this case since it gradually dies out, with having a short vocal like this there is not much overlapping going on. Another method to make these delays is using the multitap of the rv7000.

Finding the correct tone height

Finding the correct tone height on where the vocal is playing is something I do by feeling. Since I personally have got like 19 years of experience of making music, I think personally I only have to look at something and I can create something from it. That is probably also one of the reasons why it might be hard to explain in a few words how to find the exact tone height of the vocal. One minor tip I can give though is use a simple monotomic sound (like a sine form or a square form) and repeat that sound over and over again on the same tone height. Play the vocal next to it, and see if you will notice a big difference when placing the notes of the repeating sound on a different key.

Just as an example, I am using a single subtractor for this purpose:

Download RNS example file: Vocals with recycle 2
Download MP3 example file: Vocals with recycle 2

As this example file will probably sound pretty 'annoying' due to the single beeping sound comming from the subtractor,you might already notice that the beeping sound goes off-sync when playing this beep on a different tone height. You will probably have to experiment with it I guess, but this is the easiest method I can come up with to find the tone height of a single vocal. Finding the tone height of a vocal works best when the song itself hasn't been build yet. Thats one thing for sure. Just have a single sound next to a vocal works best, since the more crowded a song becomes, the harder it becomes to find the right tone height of a vocal. Since the rest of the sounds are probably in the way and can clog up the rest and you probably not really notice the difference when playing a D or an E note next to the vocals.

Vocal effects in propellerhead reason

One of the greater things you might want to focus on is using effects combined with vocals. Like I already used a RV7000 reverb engine to shape the vocal just a little bit differently, there are all kinds of different ways to shape it. If you would use short vocals (like the Lets go sample I used), my advice would be loading them up inside the NN-XT. Since there are a whole bunch of settings on which you can change the complete sound output of the vocal itself. Just to name a few: pitch effects, filter effects, lfo effects (to make it sound trembling). Just enough stuff to play with.

In the first example I am loading up a single vocal (Let the Bass kick) inside the NN-XT. I connected the Sample start to the velocity. Which can have a real drastic effect when slicing through the sample.

Download RNS example file: Vocal Effect in Propellerhead reason
Download MP3 example file: Vocal Effect in Propellerhead reason

Here's another idea that I just was working on. The main idea is using the Matrix pattern sequencer and connect the curve CV inside the NN-XT settings (such as Panning, Master level etc).

Also another effect I am using is a dynamic reverb engine to build up the momentom inside this minor sequence. Ok, I must admit, there is much more going on then I would expect... but heck, this is what you get when I am only touching reason for 20 minutes ;)

Download RNS example file: More Vocal Effect in Propellerhead reason
Download MP3 example file: More Vocal Effect in Propellerhead reason

Another method for shaping the vocals is using a vocoder unit. But as I have also made a complete tutorial about how to use the vocoder BV512 inside Propellerhead reason, I personally don't want to stand still too much at that one. You can find the complete text here : the vocoder BV512 in Reason

The real cool part about vocoding though is that you don't need to worry about the tone height that much, since the carrier input signal is making the tone height for is. So, in a way to do things right, using a vocoder unit is the easiest way to remove all the hastle and we only need to focus on the timing issues. But that one is probably easier to handle.

Anyways, to give a fresh vocoding example, I am using a star trek sample... why not ;)

Download RNS example file: Vocoding Vocals in Propellerhead reason
Download MP3 example file: Vocoding Vocals in Propellerhead reason