So... my naked voice occasionally works. Especially for emotional, poetic words which are not long and drawn out in too many places (as such a thing requires an actual singing voice). Which, unfortunately, is not the type of music I'm inclined to make because it requires better quality music, and I can't write music.
So... next Vocoder attempts will be this: I am taking a track I've made, creating a very simple WAV file of an instrument like an organ or wind instrument that creates a permanent sound (not a decaying one), and playing that in the tune of what my voice should do if it was singing to the track. Then, I will speak in time to the music, trying to alter only the volume... then rely completely on the Vocoder to turn the plain voice and the "sung" instrument into a voice that hits the rights notes. I will then take the result, add distortion and voila: Industrial quality singing voice.
What I'm trying to do is find a "formula" that I can then easily keep using across multiple tracks, perhaps modifying the amount of distortion for different tracks as appropriate. The pseudo-voice of Vexen may yet be heard! Of course, if you hear nothing more of this, then it's been a failure, and an embarrasing one!
Oh, in other news, I paid off two of my debts, I now only have two (and one of them is my student loan which I don't need to start repaying yet) remaining. I achieved this using trickery.
On the music front... my second track took some big steps towards success when I scrapped all the midi tracks (just keeping the synth work and Rebirth tracks) except a few bits, and started again. I was trying to make up for lack of tune by making it repetitive and trance-y, however it was turning out just /too/ obvious that it's produced by a complete amatuer with bad taste in music...