VO-BB - 20 YEARS OLD!

[voicejones]

Most of my work is done from my voice studio. A fair amount of it airs in the market in which I live so I get to critique myself as I watch tv. I noticed some of my stuff was not as loud as many other spots. I had been normalizing at -6 which I was told by a major studio was a standard they used. I've recently changed to a -3 normalization to make my voice as loud as the other things I hear on the air. It seems to me the producers at the TV station assembling the spots should be standardizing the levels but I think they just slap it together and air it. I'm curious as to what normalization settings are being used by all you in the trenches.
Thanks
Non Normal
JJ

[Andy]

A lot of 'em don't even know what normalize means. They just slap 'em together. I've seen it a lot where I live. It goes out all pretty and clear, winds up sounding godawful on TeeVee. For one client I even jumped on the AVID and said, "Here, kid, lemme show ya a couple of tricks."

[kgenus]

When I send dry VO, I don't normalize or eq. I do compress 3:1 going in (with a fast attack and release to reduce peaks) engaging the high-pass filter which is set at 80hz on my mic pre, but I make sure the gain is right in the pocket.

Also, and very important, low frequency content in any track may cause additional compression on the entire mix when the program content is sent through 10 levels of compression for broadcast.
_________________
Genus

[Simon Fellows]

I'm with Kevin Genus on this one. Get the level right at record if at all possible. My father was a TV & film sound/dubbing engineer so I guess it's something I got from him. This was in the pre-digital age, so if you didn't get it right and had to up the level you were going to increase tape hiss. But it's something I still hold to now. The less you have to meddle with a file after recording the better, in my opinion.

Funnily enough, last night I was having a conversation about the right db levels with my brother (who is also a film & tv sound engineer). Although things are technically different from the UK to the States, which I won't go into here, I believe the general standard level for broadcast VO in the UK (and I'm only talking broadcast) is around -10db. Perhaps Banksey has something to add to that.

And Andy, this is a sign of the times I think. If you've got someone doing the final mix on an Avid, then what you basically have is an editor, who may well know how they want something to sound, and may even have the technical ability (although, in your case, from what you said, maybe not) but who is still using an Avid. You can do a lot with an Avid but it isn't a fully equipped & configured sound studio.

All the best
Simon Fellows
_________________
Simon Fellows
http://www.simonfellows.com

[Andy]

Over here, in my experience in Television, it's always been about the pictures and not so much about the audio. So, when I can, I try to help them utilize what AVID does provide them with regard to improving audio.

[Dave]

Now that touched a nerve....

For broadcast, most of my work is TV and I have yet to hear anything approaching good quality audio on the air. On the corporate side of the vo coin...the lions share of it sounds awful as well. Only a few of my clients know anything about audio so lately I have taken the bull by the horns and the work I send them has been tweeked the way I think it should sound. So far no complaints, which tells me they are still just plugging the audio in as is.

Dave (Tried of my voice sounding like it is coming through a bucket of mud) DeHart
_________________
. If at first you don't succeed, then bomb disposal probably isn't for you.

[Philip Banks]

Processed or with no preservatives or additives, I ensure that audio leaves Portgordon in the best possible condition. What the client does with it is his decision.

-10db is broadcast standard. Can you ...sort of...almost hear it? Will do.

The curse of the Mac or the PC is that someone with an audio production package is an audio engineer and someone with Final Cut pro is a film maker. Software and hardware enables expertise it does not bestow expertise.

[dmgood]

I think you are talking about perceived loudness. Normalizing will take the peak, find the difference between it and the max you've set and raise all levels in the file by that percentage. Depending on how disparate the peak is from the rest of the file, that could leave most sections still well below the max. If there is an option for RMS rather than peak normalization it will do more of a comparison of the overall average to the max.

To boost perceived loudness you want to compress and hard limit, which squashes the peaks and boosts the whole file to the max. Downside is you lose dynamic range, whispers, off mic utterances versus hard sell, but usually is not an issue. The question is how much hard limiting, compression should be done by you versus the broadcast chain.

There are lots of posts about this in rec.audio.pro newsgroup. Do a search in Google groups on perceived loudness and look for posts by Mike Rivers.

[schaer]

Here is a very interesting audio on this very topic: http://www.westpark-studios.de/ then click on "FRAUEN", next screen click on the button "weiter", next screen choose "DSL" if you you have highspeed. Next screen: click on "KOSTPROBEN" and choose the fourth option down "LAUTSTÄRKE IM WERBEBLOCK" (the long one), next screen click on the arrow to the right, next choose "LAUTSTÄRKE IM WERBEBLOCK / ENGLISCH".

Sorry, kind of complicated to get there.... but worthwhile.

Best,
Bernard
_________________
BuyMyVoice.com
Bernard Schaer - German/English Bilingual Voice
Available via SourceConnect
SaVoa.org No. 09023

[Bill]

Okay, what I'm hearing here is -10db, I hear other places -20db for DV, (video) how do we reconcile the two?

on my Avid DV I have it set for -12db why? to be honest I haven't a clue :shock: (I finally decided to delve deeper into the system and found that is what I had set for default, while Avid's default is -20) I just want to avoid that gawd awful digital clipping. I keep catching myself trying to go to 0.

But in reality once its captured on DvCam tape its all 1's and 0's so the key is on capture, so you have to mess with less later.

This all came up recently when my closed caption vendor brought up audio levels.

anyone help clear the fog?

[Deirdre]

[kgenus]

LOL, might be Banksey...
_________________
Genus

[nick reed]

Hey Bernie,

That was a pretty "kuhl" link... funnylauffen too!

Really, thanks, I enjoyed it.

Nick

[kgenus]

I'm hoping this might dispel the myth for some people, I’m telling this from *my* perspective with the assumption it’s *generally* the same for anyone producing in broadcast.

I receive audio from talent and drop it into ProTools, setting up a VO sub mix to render an initial VO mix track. This process includes cleaning up the audio and adjusting it so the timing is right. ProTools has some really cool built in tools for this. When the VO sub mix has been rendered, I move on the SFX and music which takes longer than I would like it to…. When that process is done, I run everything through external processors (compression, EQ and limiting) and render the final file at -10db.

After that process, I drop the track into Winamp with the Sound Solution plugin active, set for the Optimod 8400 – the standard broadcast processor many stations use (see below).


Sound Solution Plugin with Optimod 8400 FM config

As you can see in the image above, at -10db, the output is near 0db. The entire mix is at -10db, not the audio I received from the voiceover talent. When the talent begins modifying the audio by processing it themselves, it just makes my job that much more difficult, which is why I never send my audio out processed.

For those of you who are adventurous and are using the Microsoft Windows operating system and Winamp, the Sound Solution plugin is freely available (see link above). You may hate what it does to your audio, you may love what it does, whatever the case, you'll have a better idea how it will sound on-air.

Hope this helps (and sorry for such a large image!).

Kevin
_________________
Genus

[anthonyVO]

Be very careful with expecting too much from normalization - like mentioned before, it won't do anything for perceived loudness - especially with audio that already has a big dynamic range.
Compress, Limit, and Limit some more if you'r looking for loudness.
But, again, be careful, because every video editor I've dealt with wants different levels. Most didn't know or care much about audio and audio mastering, so I always asked.
For the radio station spots, I know that their production departments don't touch the levels so alot of my clients always compliment me on their spots standing out - I kill the dynamics on the hard-sell spots sorry - it's not something I'm proud of.

In a nutshell - always create a dialogue with the end party and find out what THEY'RE looking for in terms of your v-o and follow suit. You'll find that although there are many variations, your usual settings will become a "standard preset" of sorts for you and a pretty happy compromise if you're too busy to dedicate too much time to each final audio file.

-Anthony