For those of you who may not know what is the VoiceProcessor, here are some words about it.
Imagine a real synthesizer built around a basic structure which is duplicated n times. That structure may be a sub-assembly capable of producing the sound of a single voice. Each time a new voice is started (following a note on for instance) then one of these basic structures is chosen to produce the sound for the new voice. Each sub-assembly - or board - has a set of settings that enables to generate different sounds for each boards. The VoiceProcessor is a PATCH module that emulates that kind of real synthesizer's structure. This module emulates a machine using 8 boards, and it provides 8 modulation outputs that can be connected to any compatible input of any module within the patch. The level of these modulations are programmable thru the UI of the module for each board and each output.
When a note on is processed, a "board" is chosen according to some algorithms (that you can choose in the UI: backward and forward circular, random...). The level of the 8 modulations will vary according the chosen board, and therefore the sound generated by the voice will also vary with these modulations.