You will not I

This is a (part of a) composition that I am working on. It consists of some 100 words that are recited, first by Henning and then doubled by myself.

First stage
At first there was Henning's recording which I played through my monitor and re-recorded while moving the microphone in front of it. That way the original recording became sort randomly flawed. I did this in four goes and layered them.

Second stage
Then I recited the lines in as much as possible the same intonation and speed as Henning and layered two of those recordings over that. Because it's impossible to do this exactly another blend was mixed into the original.

Third stage
I rendered the mix and treated it with Smartelectronix Scrubber. This vst allows you to control the play of a file (needs to be 44khz/16bit/stereo, btw). It is the most versatile and controllable granulation plugin that I know. Certainly in the freeware world. The result that I wanted (and got) is that each word (or short sentence) sort of rolls in, has a more stable middle and then rolls out again. The anomalies that scrubber produces are a bonus in the mix, really. :)

Fourth stage
Back to Reaper and start a new mix. I blend the Scrubbermix with the original ones in order to emphasize the middle part of each word.

Fifth stage
I make a live recording in my studio, using binaural mics which I move through the room and the hallway leading to the studio. This live recording is layered into the mix to fill in the gaps between the words. Those gaps vary between 1 and 6 seconds. They are bassy and zoomy because of the Scrubber anomalies and the soft spatial recording is an interesting addition.

The piece is not yet finished because the development is too weak, so now I will start working on that. It is supposed to become part of a bigger whole where the whole poem is recited from start to beginning in a few phases. I'll keep you updated with further progression.