VO-BB - 20 YEARS OLD!

[georgethetech]

So, how is a DeBreath tool different than a properly tuned downward expander or noise gate?

When I dial in the AU Dynamics effect that comes with Macs with the proper amount of ratio, it seems to debreath the track very nicely.
_________________
If it sounds good, it is good.
George Whittam
GeorgeThe.Tech
424-226-8528
VOBS.TV Co-host
TheProAudioSuite.com Co-host
TriBooth.com Co-founder

[D Voice]

[Mike Sommer]

Normalizing has nothing to do with DC offset. DC off set is when the signal has fallen off the baseline. The result is a loss of head room.
The first waveform shows the signal off the bass line by about 3 dB, thusly a loss of 3 dB of head room



What normalizing does is turn the volume up on the entire signal. It's like reaching over and turning up the volume knob up on the Hi-Fi, but in the Digital domain. Remember 0 dBfs is the absolute limit, i.e. there is no head room above 0.

So now if you start making adjustments to a track that has been normalized to 0 or even -3dbfs, you run moor of a chance of clipping the output. You can clip and un-normalized track too, so watch you output volume. This is way i say if your peaks are at or around -6 dBsf, you are right on the money.


SO, edit and contour your tracks before you normalize. Normalizing should be the last process of a final mix. Even then you should only normalize to -3 dBSF, never to 0.
_________________
The Blog:
http://voiceoveraudio.blogspot.com/

Acoustics are counter-intuitive. If one thing is certain about acoustics, it is that if anything seems obvious it is probably wrong.

[Jeffrey Kafer]

normalization should be used sparingly if at all. Most often I see files normalized to -.1db just because they want it to be loud. The problem with that is you increase your noise floor, too.
_________________
Jeff
http://JeffreyKafer.com
Voice-overload Web comic: http://voice-overload.com

[D Voice]

[Mike Sommer]

Yes a lot of plugins do have a DC offset correction. Look at your wave form. Is it offset? If it is, correct it. I personally have never had a problem with DC off set, and in some programs that can be corrected "adjusted" in the program its self. Because that is often a problem of the program and will translate to a loss of head room during recording. So if you see DC offset, fix it first before you start recording.

You can of course normalize down. But if you've clipped at 0 dBfs during recording, the damage has been done and there is nothing you can do to correct it. This is why the best practice is to record with proper levels (for me that is peaks at about -6dBfs). And Normalizing adds no artifacts to the recording.

If you set your levels correctly before you start recording, you shouldn't need to normalize. ( This may include using a compressor in your signal chain -set with a very modest compression- to control peaks. If done correctly and properly no one will know.)

Of course things happen and you may need to apply normalization.
But I wouldn't make "normalizing" my first procedure after recording. As you suggest the good thing about it is that it can make a track louder without the artifacts of digital compression, but the bad thing is that it makes the entire track louder. That is to say, if your noise floor is low or not low enough, you will hear it after normalization. This is way I say you want to do any of the changes to the track, such as noise suppression before you normalize up- which is what most people are doing when normalizing.

Also, normalization is often a destructive effect, while simple volume changes are not. If your DAW doesn't contain a very good "undo feature," normalization is often a one-way street.

"Audio normalization or "peak normalization" is the process of uniformly increasing (or decreasing) the amplitude of an entire audio signal so that the resulting peak amplitude matches a desired target (the norm). Where as dynamics compression, which applies varying levels of gain over a recording to fit the level within a minimum and maximum range based upon the selected threshold. "

[D Voice]

Thanks Mike. I guess it depends on how destructive the normalization process actually is.

Actually, of late I have been recording more in the -20 to -10dB range to be on the safe side (I usually set in one room, then record in the other), and to keep from pushing the preamps to hard,

So lately I have been normalizing UP (after the HPFF, removing any clicks and pops of course). As you say, that raises the noise floor, but there again, I figured this would make it somehow "easier" for the Noise Reduction analysis.

I suppose I have been thinking about normalization as a sort of calibration,

[Mike Sommer]

[georgethetech]

I find the easiest workflow is after recording (with peaks -12 to -6), normalize to -3db, apply a Stack in Twisted Wave with HPF, Parametric EQ, Dynamics (gate & comp), and normalize back to -3db again. This way you don't have to monkey with threshold settings every time since the levels coming in are consistent thanks to normalizing. Then after the Stack, normalize again to make up gain for total consistency.

This technique has made many a client of mine very happy...

Problem is, some software (such as Soundbooth) has a fixed normalize level of -.1db, which may be problematic. I'm not aware of a case in which normalizing to a value less than 0db is destructive any more than a volume adjustment is destructive. The process simply scans the selected waveform, finds the peak level, and bases the change of volume from that peak.

If I am wrong, someone will surely correct me.

Oh, and how is de-breath different from a noise-gate? Still curious about that one.
_________________
If it sounds good, it is good.
George Whittam
GeorgeThe.Tech
424-226-8528
VOBS.TV Co-host
TheProAudioSuite.com Co-host
TriBooth.com Co-founder

[Mike Sommer]

[captain54]

From waves.com

Unlike classic noise-gate type processors based solely on energy threshold, DeBreath uses a unique template-matching algorithm to detect breath segments and separate them from the signal. The breath can then be controlled without affecting the rest of the signal.

DeBreath lets you control the amount of gain reduction applied to the detected breaths and to add room tone where breaths were removed or attenuated, thus avoiding unnatural holes in the track. For better control, you can split the audio into two paths: the Vocal path and the Breath path. The sum of these two paths will always equal the original source, regardless of DeBreath settings. This way, you can apply different processing to the vocals than to the breaths. For example, you might compress only the voice path, leaving the breaths quieter, more open and natural. Or, you might choose to darken the removed breaths.

[asnively]

For reasons unknown, DeBreath is showing up in my Effects menu (Audition 3, Win 7) as "unsupported."




I've uninstalled, rebooted, reinstalled, uninstalled, reinstalled, rebooted, uninstalled, reinstalled, rebooted, rebooted, rebooted; trashed my Audition preferences (and reinstalled); unplugged the iLok from the USB hub and stuck it straight into the computer; stood up, sat down, fought, fought, fought; sacrificed a goat.

I'm stuck.
_________________
the Amy Snively family of brands for all your branded thing needs.
Amy Snively
Faff Camp
FaffCon
TalkerTees

[Mike Sommer]

+1 the Captain's information.

I missed the question. But in basic simple terms I think of a Noise Gate like a garden gate. When you want to pass through the gate swings open, when not, it's closed.

De-Breath is more like a bouncer at a night club, sometimes you get in, sometimes you don't.


Amy, it sounds like there are some fragments leftover from that plugin still lingering. You may need to dig through some folders to find those residual tidbits.
_________________
The Blog:
http://voiceoveraudio.blogspot.com/

Acoustics are counter-intuitive. If one thing is certain about acoustics, it is that if anything seems obvious it is probably wrong.

[asnively]

What am I looking for? Where do I look?
_________________
the Amy Snively family of brands for all your branded thing needs.
Amy Snively
Faff Camp
FaffCon
TalkerTees

[Brad Venable]

Mac or PC, Amy?
_________________
Brad Venable
The Voiceover Superhero!
"Saving the world thirty seconds at a time."