How to approach sound levels

Deuceofspades

Deuceofspades

Veteran
I need to work on the sound on my short, next.

I need guidance please :).

What is the best way to approach it. Each scene is its own project, then I have the master project.

1- What is the target level of finished film on the meter? -12 db? Higher?

2- Is it best to get each scene as close to possible to that level and then do final twicks in master project?

3- What sound processors do you usually use (noise gates, EQ settings etc...)

4- I have the sound of thunder in one scene and it always peg in the red, unless it so low in the mix that you can barely hear it. How can I make it sound loud (for emotional impact) without distorting?

Thank you for your help and advice,
 
Deuceofspades said:
I need to work on the sound on my short, next.

I need guidance please :).

What is the best way to approach it. Each scene is its own project, then I have the master project.

1- What is the target level of finished film on the meter? -12 db? Higher?

2- Is it best to get each scene as close to possible to that level and then do final twicks in master project?

3- What sound processors do you usually use (noise gates, EQ settings etc...)

4- I have the sound of thunder in one scene and it always peg in the red, unless it so low in the mix that you can barely hear it. How can I make it sound loud (for emotional impact) without distorting?

Thank you for your help and advice,
Click to expand...

I like to know too. There seem to be several ways to control levels.
 
1 -- Generally, keep the peaking averaging around -12, yes. Don't let your louder peaks go much higher than -6 or so -- at -3 you're at serious risk of clipping. Don't let dialogue fall too far below -18 or it'll be hard to hear.

2 -- That's a legitimate approach -- though if you've kept consistent levels when editing your scenes, you shouldn't have to do too much for the full project.

3 -- As little as possible. The place to get great sound is recording it well in the field. If you do that, keeping your noise floor low and your signal strength high, keeping a consistent level should give you pretty clean, clear sound (which is the goal of recording audio in the field). Noise gates and EQ affect your dialogue, and if you're trying to get rid of noise, etc., you're going to be cutting away at your actors' voices.

However . . .

4 -- You can use EQ, etc., for effect. In this case, a way you can go is to use EQ to raise the volume of the lower frequencies. It shouldn't affect overall volume that much, but it should give you a pretty satisfying rumble.

Now, you don't want to be recording this thunder along with your dialogue, of course. Sound FX are meant to be added later; on set, you just want to be getting the dialogue. If it's convenient to record some sound for foley FX, do it separately from the dialogue.
 
Hi Faith,

> 1- What is the target level of finished film on the meter? -12 db? Higher?

-12dB is standard, but that assumes that there will be other peaks you
can't see that might go higher. You can actually finalize your mix louder
if you use something like the Waves L2 plug-in, but for general purposes,
-12dB is good.


> 2- Is it best to get each scene as close to possible to that level and then do final
> twicks in master project?

In general, you want to get good clean sound when you shoot, and then play
with the levels creatively later.

Since you're talking about a large project made up of several scenes,
you might try a somewhat different approach. Ideally you would find
a couple of key moments where the sound is loud and good and works
for you on a variety of playback hardware (headphones, small speakers,
big speakers, mono, stereo, etc.). With those moments as your references,
you can tweak the rest of the film. By doing the tweaking in each scene--
instead of in the master project--you might have better results because
you will presumably have all the sound tracks available to help you finesse
the tweak.


3- What sound processors do you usually use (noise gates, EQ settings etc...)

This is a huge question :)

Compression is useful for evening out the levels.
EQ is like color correction to balance out lows and highs.
Noise gates can help cut out noise in between sounds,
but they can often do more harm than good. The key
is to listen on many different sound systems to define
what the problems are, then you could post specific
questions. Like the following:


4- I have the sound of thunder in one scene and it always peg in the red, unless it so low in the mix that you can barely hear it. How can I make it sound loud (for emotional impact) without distorting?

It's possible that the problem is because the extreme low frequencies
are super hot, even though you may not hear them on the speakers
you are using. In that case, you might try using EQ to reduce the
frequencies below, say, 80Hz, and boost the frequencies at about 100-150.
That way you could retain the sensation of bass without blowing
out your levels with something that a lot of people won't hear.

The other trick is to compress or limit the sound. A good limiter
(like Waves L2) can make you sound a lot louder and guarantee
you don't go into the red.

Finally, there is the idea of relative mixing. If you keep your dialogue
and other stuff somewhat low (for example, -16 instead of -12),
then bringing in the thunder at -9 would make it seem quite loud.
This works fine, but make sure you don't pull down the dialogue too
much.

Finally, you can try stuff like adding lightning flashes or camera shake
to make the scene more intense without raising the levels.


Best wishes,
Les
 
Waves L3 Ultramaximizer or other limiter. Put one on the final audio track (at about -3). Create audio busses for dialogue, sound fx, and background noise with a limiter on each one. Dialogue needs to be very heavily compressed, sound effects less so, and ambience lesser still.

With individual busses you can mix the levels so that the sound effects peak over the dialogue when they have to, but the dialogue remains clear throughout.

This is a very simple explanation. I usually have a very long chain of noise gates, compressors/expanders, EQ, and limiters in order to get good audio. Each project is different.
 
Peak level is not average level.
You need a VU meter to see the average level. This level is the reference level for the actor's voices, the music and the SFX.
 
pmark23 said:
Waves L3 Ultramaximizer or other limiter. Put one on the final audio track (at about -3). Create audio busses for dialogue, sound fx, and background noise with a limiter on each one. Dialogue needs to be very heavily compressed, sound effects less so, and ambience lesser still.

With individual busses you can mix the levels so that the sound effects peak over the dialogue when they have to, but the dialogue remains clear throughout.
Click to expand...

thank you for the great inputs everyone. I have a question regarding above pmark23:

1- "create busses". I guess I have not gotten that far yet in my Vegas uses and do not know how to do it. Are the busses created in Vegas? Or outside? Can you give more details please so we know how to proceed.

2- Maximisers... Ah yes I used to always use them on vocals. I just remembered I have a narly plug in for my recording studio. I should try it on the film dialogues. meanwhile, can I have more details about Waves L2 and L3? Are they part of Vegas built in audio plugins or third party plug ins and if so how much do they cost?
 
Right-click in the window were you see the audio levels, and select "insert bus". Another fader/meter will appear.

Now pick the audio tracks that you want to go to that bus (for example, all dialogue tracks, or all SFX tracks), and click the square button near the top on the right of the red button. It's at "Master" now, so select the "A" for the first bus.

Now your audio from those tracks you switched to "A" will go through the bus, and then back out through the "Master" output. So you can drop an effect or compressor/limiter on the "A" buss and it will only affect those tracks. Plus you can now control only those levels and have greater control over the mix.

What makes Vegas great is that you have fine control over audio. You can drop an effect on a track (which I use for filtering, noise-reducing, or gating dialogue), then group like tracks into groups to compress them together so they have the same levels and can be mixed more easily with the other audio (backgrounds, fx, etc), and then finally an gentle compression and EQ on the final mix to blend everything together so nothing is too loud or too soft.

You can even route busses to busses if you want to get really nutty.

Waves plugins are third-party. The ones I use the most are L3 which mostly goes on the final output, R-Vox, which gates and compresses individual tracks, and R-Channel if I need finer control.

Vegas has it's own built-in effects, but Waves has better compressors and especially limiters. The Vegas EQ is fine and I use it a lot.

Vegas started life as a multitrack audio editor, and it's still quite powerful. I use it quite often for audio-only projects when I don't need MIDI.
 
You must log in or register to reply here.
Back
Top