Hi all, So, I'm side chaining two mic inputs from my 8pre into two instances of the Evoc20. (One input for each Evoc20). I need two inputs because I may have two vocalists speaking/singing live into the vocoder at the same time. (One student may not be able to cover the part by him/herself, so .... two heads are better than one). This routing seems to be very prone to feedback (very hi frequency). I'm pretty sure that the feedback is coming from the Evoc CS, and not the input CS, because if I mute one input CS, the feedback is still about the same. If I mute an Evoc CS, the feedback is greatly reduced (but not necessarily eliminated). This is when I'm just speaking into one mic (the unmuted one) as a test. What can I do to eliminate the feedback but still have enough overall gain for the vocoded sound to be heard?

By any chance are you using software monitoring in Logic plus direct monitoring (via CueMix) from your 8pre?

Hey, Peter. Thanks for the reply. No, CueMix is not involved.

Midi information has already been recorded onto the Evoc tracks. Input monitoring is on for the live audio tracks used for the two mics going through the two 8pre inputs. The live vocals going through those mics are side chained into the Evoc, and that triggers the midi notes already on the tracks.

What about software monitoring in Logic? Is that on or off?

Input monitoring is on for the audio tracks. No software instruments are being played live at this point.

agagag, Could you post a copy of your Logic project? If so, remove all regions and audio files (from the bin), zip it, and post it on a server. Reason I'm asking is because I'm curious to see your signal routing. For this one I think it would be easier to understand what's going on if I had it in front of me rather than trying to decipher the routing using, you know, words and stuff LOL!

Hi Peter, I was able to do some work on the feedback issue today, and found that a lot of the problem was coming from the fact that I had the U/V Detection in the vocoder way up. I initially did this to help make the words more intelligible and did not have the feedback issue because the vocal track was pre-recorded. The group of students I'm working with is now transitioning to saying (or singing) the words live into the vocoded region as part of a live performance, thus the need to side chain the live inputs into the vocoder. I turned the U/V Detection down, and it's much better. I think if I have some more time to work with the group on stage, and hear everything else that's going on live, I will be able to get it pretty good. If I have any more problems in that area, I'll try to post the project to you, as you generously requested. Much thanks, Adam