Wave Hammer and RMS levels

1marcus4 wrote on 11/8/2009, 8:56 AM

If you use the Wave Hammer to raise the power levels for voice (as an example), how do you fine tune the settings to target the RMS level you are shooting for, i.e. -12 db for voice?

Mark

Comments

John_Cline wrote on 11/8/2009, 11:27 AM

Vegas does not have a way of measuring the RMS level of an audio track.

Typically, I set all my projects to an RMS "average" loudness using Sound Forge. I normally set it to -20 for a project that consists primarily of voice or -15db is it's primarily music.

First, I go to "Process" > "Normalize" and scan the levels, noting the RMS level. Let's say it's -27.4db with a few peaks to -2db, I subtract 20 (or 15 for music) from that. Let's say it's a voice project, so 27.4 minus 20 = 7.4. I then get into the Waves L2 Maximizer and set the threshold for -7.4db and let it do it's thing. (Setting the threshold in Wavehammer accomplishes the same thing.) This raises the loudness to -20db and lops off a few peaks in the process. I end up with a program that has an average loudness level of -20db. In case the RMS level has ended up louder than -20db, I merely use the volume process and turn it down.

Of course, as far as the level relationships between VO, nat sound, effects and music, mixing is an art and I have simply used my ears and used the meters to make sure I haven't gone over 0db at any point in the process.

-12db for voice would be way too hot.

John

farss wrote on 11/8/2009, 12:51 PM

"-12db for voice would be way too hot."

Isn't there an ATSC standard for this, -27dB ?
-12dB RMS is into the realm of very heavy heavy metal.
A voice compressed that much would be diabolic to listen to.

I'd suggest that using a compressor to set RMS levels for speech is the wrong way to go about the task. Not that compressing speech a little is a bad move but it's not the way to set the level of it and most certainly not using the RMS mode of a compressor.

Bob.

1marcus4 wrote on 11/8/2009, 2:40 PM

Thanks all. That helps a little. I'm a 6 year Vegas user but only a 2 month Sound Forge user. So I am lacking in audio expertise at this juncture.

Okay, so all that makes sense but some things DON'T make sense.

For example, the SF Normalize 'speech' preset is set to -10 db, -16 for music. And for Wave Hammer the threshold is set to -10 db for 'voice', -6 for 'master'.

My problem is that I have both audio tracks ONLY in SF and audio tracks w/video in Vegas (both destined for the iphone) that have a peak level between 0.00 and -1.0, but average RMS levels around -28 which sounds WAY too low. And I'm just not sure how to bring the power levels up without destroying the top end.

These are discourses and lectures recorded in a classroom or small hall / theater setting so the dynamic range tends to be greater than the 14 db I understand is typically targeted.

Just looking for clues to how to proceed.
Thank you again for your help.

Mark

1marcus4 wrote on 11/8/2009, 2:47 PM

BTW, that -27 db standard I believe is for encoding AC-3 audio for movies destined for dvd. Has to do with dialog normalization. I'm not 100% sure though.

Mark

John_Cline wrote on 11/8/2009, 3:05 PM

Audio is a very difficult thing to do well. The method I described above is just a suggestion for finalizing the audio level of the entire video. There are a whole bunch of other steps that must be performed correctly before you get to the finalizing stage. Mixing by the numbers is rarely a good idea as most of the things that determine the quality and feel of the audio simply cannot be measured or quantified.

That said, I'm going to give you some numbers to mix by for this particular project. You say the RMS level of the program is -28db. Get into Wave Hammer and bypass the compressor, then go to the "Volume Maximizer" tab and set the threshold for "-8db" (-28db minus the target level of -20db is -8db) and set the output level to "-0.3" (just to leave a tiny bit of headroom for D/A overshoot) You can experiment with the "longer look-ahead" checkbox and you can try setting the release time to less than 100ms. Even 10ms might not be unreasonable depending on the material. The artifacts of a really fast limiter are usually less audible than a compressor. Also, there is no "typical dynamic range."

1marcus4 wrote on 11/8/2009, 7:01 PM

Yep. I've read enough about audio over the years to know what I don't know.

Thank you John for the tips.
I'm going to tinker around with some files this evening.

Mark