Technique for ducking

I did a search on this and came up with nothing so I thought I'd ask.

If I have one track of mixed down music and another track of spread out
voice messages, how would I set up a compressor to reduce the
music track when the voice is being heard?

Thanks for the help.

RC

ORIGINAL: rcastp

I did a search on this and came up with nothing so I thought I'd ask.

If I have one track of mixed down music and another track of spread out
voice messages, how would I set up a compressor to reduce the
music track when the voice is being heard?

Thanks for the help.

RC

Pipe a copy of the voice track into the side chain so it will trigger the threshold for clamping down on the music. You can even have the "copy" of the voice track for the side chain be real loud so it works better but doesn't get included in the final mix that way, use only the first copy of the voice track to do the actual mix. On the compressor settings for the music track set your ratio high and tune the threshold till it is reached easily by the side chain input.

OK, the above is the "automated" way to do it, if you want control over each one you can use a clip gain envelope to bring down the music track. It's more work if it's a long program but you get better control. The compressor / sidechain thing will do the same thing every time. With an envelope you can adjust each duck seperately. The other advantage of using an envelope is that you can make sure the music stays ducked till an entire multi-word phrase is done, an automatic ducker might pop the music back up between words.

ORIGINAL: rcastp

Thanks for the help.

Normally I would use a volume envelope, but because I'm working
on a music on hold project, I wanted to let it duck the music automatically.

I should have been more specific - as I could and may well have to use
the sidechain on my hardware compressor - I was trying to find out if
there was a software / plugin way of doing this.

Thanks,
RC

I think there are sidechainable compressor plugins now but I don't have any yet so I can't comment on them. However, they should work exactly like the hardware type.

If you have SONAR 7 it now supports traditional side chaining via key put on the compressor. So you'd just put a compressor on the music track and feed the voice track to the key input. For SONAR 6 or earlier, you can do a more home grown method where you send them both to a stereo bus, one panned hard left and one panned hard right, and set up the compressor to key on the left channel, then pan the buss hard right so that you only hear the right side (which has now been ducked) and feed it back to another buss so that you can pan it back to where you want it again. I believe I'm remembering that correctly.