But I don't go too crazy because eventually I'll have to mix it with other elements that sit in the same freq bands. It is better for the neighbors tho.Īlso, I detune the z3ta+ oscillators a bit, set at them at different octaves or subtly blend in a fifth for more texture. One thing I've discovered is that a harsh whisper, close to the mic, works much better than imparting any tone or actual voice (which seems to blow out the mids and just sounds like a garbled synth), but that could be cause I'm doing it wrong. I know some of this is specific to Vocodex, but I'm looking for a broader understanding of how to get more intelligible vocoding. Haven't had much success messing with other parameters. I do a L/R encoding, with my (compressed) voice as modulator and a mix of square, saw, and sine waves from z3ta+ as the carrier. I'm using Image-Line Vocodex, using around 40-70 bands, a shallow 2x chorus, order 2, SG at full, and 100% wet. Lately I've been trying to vocode my voice, but haven't gotten very intelligible results.
