I will try the sawtooth and some other things. Meanwhile, I heard back from Ableton Support and they at least helped me with the MIDI track volume problem somewhat. The support guy understood that part of my problem, but not the rest, so I'm going to post most of the text of my answer back below. Maybe someone here (even if it's only you) will see something I didn't quite clarify before and an answer will come. If not, I'm giving up! I can't even find anyone else on the 'net asking this question, so it might be the question itself that's flawed, I just don't know. Thanks again.
From my e-mail back to Ableton:
"Thanks for responding to my question and sending back my project with the effect rack.
Adding the compressor does help some. It is now easier to hear the carrier as it's affected by my mic vocal. I played the synth loop you provided so I wouldn't have to keep pressing keys -- thanks for that. I'm new to side chains and chains in general, but I'm trying to learn, and I watched the [sidechain] tutorial you linked me to as well. Thanks!
My problem still remains, though, of still hearing the dry signal while using the vocoder. Maybe my lack of knowledge is preventing me from understanding the nature of how this works, and I'm beginning to wonder if dry signal (and hearing it) is necessary to do the actual vocoding. As I said before, in Houston Singletary's videos (Part 1 and 2) about the vocoder, he explains everything *except* how he suddenly completely eliminates the dry modulator signal for a few seconds! Now, I don't think I'll ever need to do this, but just a hypothetical -- what if I were using Ableton's Vocoder in a live performance? Would the audience hear the dry signal mixed with the vocoder output?
Let me see if I can make this better understood, because I think it's a communication problem on *my* end, mostly due again to my lack of experience or knowledge. First, I know that I can get the result I want by recording what I'm doing with the vocoder and end up with just the vocoded material (no actual voice in it). I can also record audio and synth separately and then use the vocoder, etc., however that's done (I'm sure that part's easy). But what I want to do instead is record the vocal while playing the synth. It is difficult to decide, for example, how much depth to add or experiment with the formant, frequencies, etc., if my voice is over all of that, even if it's less than it was before.
So, as clearly as I can say it: I want to eliminate my voice from my headphones/awareness as entirely as possible while adjusting the vocoder's settings, effects, etc., while playing the carrier into it. I know I will obviously still hear my voice naturally, but I'm talking about the signal actually going through the mic which ends up in my headphones along with the signal I am trying to hear. I hope that made sense. I'm looking for a way to speak into the mic, and the meter picks it up as necessary to modulate the carrier coming into the vocoder, but I don't want to *hear* it while it does that!
If I've overlooked something in the effects chain you included that would do this, then let me apologize. I tried adjusting the overall gain, the threshold on the compressor (unwanted feedback happened almost immediately when I did that, so I left it where it was). Could my problem be the mic pre-amp I'm using? My mic level is not going anywhere near the red, so I just don't know. In fact, when I first opened the project you sent back and turned the mic on, muted track 2 and played the synth, while speaking *softly* into the mic, there was echo, which always seems to come right before mega-red feedback loop from the combined signals.
Thanks so much for your help. If there is no way to do what I'm trying to do, even knowing that there is no way to do it will help, so I can get beyond it and try the other alternative (prerecord the audio). But I know people do this live, so I figure there *must* be a way to eliminate the voice."