Making a Rising Section with Vocal Stuttering, in REAPER
4 min readCreating a riser from a vocal sample is a really easy thing to do in most DAWs, it has many routes and methods of execution but the simplest is repeating a vocal snippet, which gets smaller at measured intervals. If you are using REAPER, learn where your “s” key is as you will need it (for those that don’t know, it’s the split item function). As long as you are happy splitting and copying items all over the place, this is relatively easy.
This tutorial will be based loosely on a previous one, but I will try and supplement more detail and make this more REAPER specific.
Step 1 is to get a rough idea of your track. In the picture below, I have a drum sample, a bass sample and a vocal sample. (These don’t have to be samples, they could be recordings or synthesised.)
This is just the basic outline to work from, no doubt plenty will change as you carve the sound. In my example photos, you will notice halfway through that the drum sample changed. Please zoom into the photos as there is text and detail that may aid the process.
Step 2 is where you solo your vocal track and look for the part you want to stutter. This may be the last word of a verse before the chorus, or it may be something else. You can play it over more vocals or it can be by itself.
What I do recommend, however, is you pick a word at the end of a phrase where, musically, it is more likely to have the intonation and cadence that wants to resolve into the chorus. Think almost like in a question, where the last word is upwards inflected, with the expectation of an answer after.
In the above picture, you will see I have highlighted the section that I like. I have made sure the snap function is enabled so that when I split it, it is an exact division of the beat. This is important as when you copy and paste an item several times, the small inaccuracies add up.
Step 3 shows the rest of the vocal track deleted and the selected item brought to the start of the build. Remember it is just in the context of this section of the song so you obviously don’t have to delete the whole track.
I’d like to point out the purple line just now, on the drum track. This is just a high pass filter, you will notice the bass is getting removed until the bass-line kicks in. This is typical of most builds, where you remove all the bass so it is perceived louder when it does come in.
Step 4. So you will notice that the above image shows the selected sound in two parts (divided into two beats), for now, I just glued them together and copy/pasted them along the length of the build. See below.
Step 5: I noticed that because the audio was split, it starts off at a peak. When listening through, it was quite hard hitting in a bad way. I decided it wasn’t good for the sound and so added a fade to the start of the item, see below.
This is a good point to decide if you want to further decrease your beat length, I decided not to just for simplicity but it adds nothing more difficult than a few more minutes of your time.
Step 6 was to get the sound to actually start rising. If we increase the pitch with automation, we will make the vocal go in and out of tune. We don’t want that, which is why we edit the formant instead. It will mean the frequencies rise while the pitch itself is unchanged.
I used ReaPitch to increase the formant and automated the formant control. If you notice, the slope isn’t linear. I decided to set it as a “fast-end” curve. This was to maintain the initial integrity of the sound before it suddenly increases in formant right at the end of the build.
Step 7 is to increase the feel of it rising by adding other effects to it. I decided to add ReaDelay on a ping-pong setting and automated the wet mix from 0% – 100% as the build progressed. Increasing delays like this add a lot of intensity to the sound as well as blurring it up, they are a good way to build to a drop, in conjunction with other effects.
Step 8 was a spontaneous decision, the sound seemed very empty. This was partly due to the lack of other instruments and the limited bandwidth of the vocal frequencies. I decided to make another track and take the second half of the sample, this was repeated to the end of the build and they were glued together.
“Shift” + “w” brings up the stretch markers and so you can get a double the speed of the playback. This tool was used to make it an increase in energy and decreasing in note length. I just copied and pasted the effects chain from the original vocal track, with the exception of dropping the pitch down one octave and fiddling with the formant until it sounded nice.
Step 9 was adding more building effects to the mix. I decided to focus on the second vocal track that was just created. This track is designed to sit below the main one in the mix so we can make it sound a bit rougher. I decided a ring modulator would sound cool. I set it and automated a rising frequency, in the same way the formant and delay were automated. This produced a rising effect, which sounds a lot rougher than other methods but in a nice way.
Step 10 was simply to mix the resulting sounds and there you have it.
