finally: an easy solution for audio ducking

Jeff Waters wrote on 7/4/2008, 10:55 AM
Gang,

Maybe there's something new for Vegas on this (please tell me if there is!) but a couple years ago I was looking for a good solution to automatically duck a music bed behind spoken word.

The best I could find at the time was the BlueCat envelope tool and Excalibur. The BlueCat tool never seemed to work quite right for me, though. And besides, it was a few too many steps for my taste.

I just discovered that the new 1.3 beta version of Audacity (which is nowhere near as nice as my beloved Vegas, but is absolutely free) has a new Auto-Duck effect. It works beautifully!

So, my workflow will be to do all my editing in Vegas. Render the music bed in one file, and the audio track in another. Import those into Audacity. Do the auto-duck effect. Then render everything to a final mp3 from Audacity.

Man, I would really really love to see the same kind of Auto-duck tool right in Vegas. Have I missed it? Is there anything new that'll do this without me having to leave Vegas?

Thanks!
Jeff

Comments

Tom Pauncz wrote on 7/4/2008, 12:15 PM
Jeff,
VASST Ultimate S Pro from the good folks at VASST can do just that, without ever leaving Vegas. The latest version behaves as an extension (as well as a script), so you can always have it available via the tabbed windows on the top left.

It isn't free however, but well worth the bucks.

Tom
TLF wrote on 7/4/2008, 12:30 PM
Audacity 1.3.5 is a nice improvement on previous versions. I've just tested the Auto Duck feature... It's OK, but not perfect. I'm sure I can make use of it.
Jeff Waters wrote on 7/4/2008, 2:10 PM
Thanks Tom,
I didn't know S3 had added that. Does the S3 ducker trigger by events (ie, the beginning or end of a clip) or by a volume threshold on the vocal track?

Does it allow you to set things like pre/post fade times on the duck?

Is it a real-time operation or does it actually change the target music bed? This is the only problem with the Audacity implementation... it actually changes the waveform, so you need to monkey with it and do a bunch of "undo's" until you get what you want. It would be nicer to be a real-time plug-in so you could hear the effects as you go. Though, I wouldn't kick it out of bed for being destructive as long as it's inside Vegas...

Jeff
Tom Pauncz wrote on 7/4/2008, 3:11 PM
Jeff,
Actually it's US Pro v4. I haven't played with it in a while, but I am reasonably sure it's not real time and I believe it's the volume threshold, but again, not sure. You could download a fully working trial and check it out though.

If Spot or JohnnyRoy look in perhaps they can chime in.

Meanwhile, I put a screenshot on my server to show you what it looks like.

Tom
Jeff Waters wrote on 7/4/2008, 4:31 PM
Hey thanks! I'm curious to find out if this creates a volume envelope or actually changes the waveform.

Spot?
John_Cline wrote on 7/4/2008, 4:57 PM
Ultimate-S, both 3 and 4, duck audio by looking at events on the VO track and then generates a volume envelope which is applied to the music track. It is customizable and totally non-destructive. It does not adjust the ducking based on the VO track exceeding a volume threshold, but the event edges. Everywhere there is an event on a track, it will duck the audio on the other track. Works really well and gives you quite a bit of control over the process.

The behavior is described here:

http://vasst.com/?v=ultimate/us3/us3_audio.htm
JohnnyRoy wrote on 7/4/2008, 5:25 PM
> Ultimate-S, both 3 and 4, duck audio by looking at events on the VO track and then generates a volume envelope which is applied to the music track.

That's correct. It is event based and does not monitor the waveform in any way. It looks at VO event edges and ducks before, during, or after the edge is encountered based on the options you have selected.

~jr
Jeff Waters wrote on 7/4/2008, 5:51 PM
Ah, thanks guys. Hmmm. I would think my vocal recording will have only 1 front edge and 1 back event edge. I think an idea ducker plug-in for Vegas will have to be triggered from volume threshold on the VO track and create a volume envelope (that could later be adjusted if necessary). 2nd best would be to destructively change the waveforms.

I'll stick with my hackneyed Vegas->Audacity workflow for now.

I imagine this would be a popular plug-in if any developers are reading this!
John_Cline wrote on 7/4/2008, 10:04 PM
It's not that difficult to go through the VO track and cut out the "dead" spaces. Most of my VO work is already done in individual, trimmed clips, so Ultimate-S works really well for me. Regardless, Ultimate-S is a tremendous value for the money.
TLF wrote on 7/4/2008, 11:15 PM
Having tweaked the settings in Audacity's Auto Duck feature, I am changing my mind - it is surprisingly good.

I tested it with th audio from a project I am working on , and it has done a fantstic job, much better than I managed.

Highly recommended.
ScorpioProd wrote on 7/4/2008, 11:40 PM
Side chain compressors are great for auto ducking.

I used to use the free SideKickV3 VST plug-in side chain compressor from Twisted Lemon to do this. And it really worked great, setting up one track's compression based on the output of another track.

Problem is, it stopped working with Vegas around Vegas 7.0c, and I've been without one since then.

I need something based on the actual audio levels of a track for my needs. I do things such as cast commentary tracks on videos, and it wouldn't be time effective to manually try to cut that track up for the other Vegas auto ducking method mentioned.

Since Vegas Pro 8 came out, I have been experimenting with another way to do this, using only the plug-ins that come with Vegas Pro 8. By using the Wave Hammer Surround plug-in with its routing feature and a mono mix of my project, I can use it to the control the level of a mono mix of my commentary track, by setting up the L/R bussing that way. That plug-in lets you do true side channel compression by letting one channel control the level of another channel.

It's kinda complicated till you do it a few times, and kinda hard to explain how to set it up, but it does seem to work.