Wicked Sequencing Alchemy for Vocal Samples

I use this simple, but extremely effective technique with a vocal part, but it can be used for any sample. From here on out I will describe the technique in reference to vocals. If your sequencer allows cutting and pasting of a sample on your track, then read on. If not, I recommend getting one that does. I prefer hardware (the MV-8800) but any contemporary sequencing software will do. Its hard to describe the effect that this produces because it depends on the sample you begin with, so bear with me as I try to explain the process and try to imagine the results.

Place the part you wish to manipulate on one track, copy it to another track directly below and then create a track below that to use as your canvas for the part. Mute the first and second tracks and unmute the third. The first track will only be used for fast access to the original in case you want to start over.

Now begin slicing small fragments from the original track and placing them in random space in the track below. Take some larger portions and do the same. You may want to repeat some slices rather than using all of the original. Continue doing this until the second part fills the same time in bars as the original.

Listen to your new part. It may take several passes to get something that really sounds impressive. If the first pass is bad, try rearranging the sonic slices. Starting from scratch might be discouraging but once you get something that really shines you will feel that the work was worth it.

The goal is to produce something melodic and rythmic that fits with the song using the human voice. You could break it into words, but my approach is to consider it a very weird instrument rather than a manipulation of vocals. Most of the time I don't sing any real words because they can be distracting. I prefer to use short bursts of screams, glottal sounds and other noises that you won't want anyone actually hearing you make. This seems to produce the best results, and as ridiculous as the original recording might have sounded I get people picking that part out of the song and telling me how cool they think it is.

A similar idea is to only remove (not moving the slices to another track) very very tiny slices from the original which sounds similar to a slicer effect. This is a more effective approach if you actually want the sample to more or less be recognizeable with respect to the original. I don't personally use this method with vocals but have with other sounds or spoken parts culled from other sources.

Good luck