Headroom, Vocal Dynamics, and Transient Peaks

S

silentnoise

Member
I was doing some sound work for a student film working as the boom op, mixer, and recordist. I found that our actors' dialogue has so much dynamic range that some dialogue would be recorded as low as -40 to -30 db, with transient peaks up to -6 and everything in between. The script was a comedy had a lot of quiet and calm lines along with full on yelling and screaming in almost every take. I tried to ride the levels somewhat, but the actors were not consistent. never mind that its hard to boom with one arm. We were also using the me66, so it was highly sensitive to the sounds coming directly out of the mouth of the actor but not the room's reflections. Apparently the school doesn't have any hypercardioids.

What is the best way to handle this situation? I wanted to allow enough headroom so as not to clip transient peaks. So how much headroom is appropriate? When are the normal levels so low that they are unusable for a post mixer? Our mixer had good preamps (shure fp33)
, and we didn't have to drive them much at all to get hot signals out of the me66, so the tracks should not be too noisy.

It was my opinion that the actors needed more training on controlling their dynamics so that we could record consistent levels throughout the scenes. I tried to explain this to the director/producer, who recommended I just leave a lot of extra headroom.
 
This is what rehearsals are for. That being said, it's really hard to one man band this kind of stuff...booming and riding levels. The more headroom your mixer has the better, but if you leave too much, you'll have really low level audio, and when that gets raised later by the re-recording mixer, you'll just get more room noise and more noise from the mixer's pre's. If it's a student film, then I'm guessing there won't be lots of high end plug-ins available to remove junk like this. Good actors can kind of control their dynamics, but the more they pull back on their dynamic range, the less natural it will feel, and they will feel restricted. It's really not their job to make sure you get good levels, just like it's not your job to make sure the actors know "what their motivation" is.
 
Your experience is fairly typical, and one to be expected in the field. Actors will speak at one level during rehearsals, and change completely during the takes, and often varying greatly from take to take and during the take. Keeping your peaks at -6 is right about where I like mine to land.

And now that I'm venturing into post production audio, I have found very very few dialog tracks that did not need level automation to smooth things out a bit. But these are only there to smooth things out a bit. The actor typically uses a very different tone when speaking softly, or whispering, then when speaking loudly and I like to capture that and use it to help tell the story.

These are the kinds of things that make production work such a challenge and the reason why we use such expensive microphones and choose the one with the appropriate directivity and method of off-axis rejection. Those dynamics are part of the magic of film dialog and communicate so much of what the actor is trying to express

silentnoise said:
I was doing some sound work for a student film working as the boom op, mixer, and recordist. I found that our actors' dialogue has so much dynamic range that some dialogue would be recorded as low as -40 to -30 db, with transient peaks up to -6 and everything in between. The script was a comedy had a lot of quiet and calm lines along with full on yelling and screaming in almost every take. I tried to ride the levels somewhat, but the actors were not consistent. never mind that its hard to boom with one arm. We were also using the me66, so it was highly sensitive to the sounds coming directly out of the mouth of the actor but not the room's reflections. Apparently the school doesn't have any hypercardioids.

What is the best way to handle this situation? I wanted to allow enough headroom so as not to clip transient peaks. So how much headroom is appropriate? When are the normal levels so low that they are unusable for a post mixer? Our mixer had good preamps (shure fp33)
, and we didn't have to drive them much at all to get hot signals out of the me66, so the tracks should not be too noisy.

It was my opinion that the actors needed more training on controlling their dynamics so that we could record consistent levels throughout the scenes. I tried to explain this to the director/producer, who recommended I just leave a lot of extra headroom.
Click to expand...
 
If you're just running the one mic, are you not running a safety channel on the recorder? Your second record channel should be about -6dB lower than the first... possibly more if you need it. That gives you a backup track in case of clipping on the primary record track, but you can also use the safety track for normal-to-loud dialog and the primary track for normal-to-soft dialog.

The added benefit here is that you don't have to ride levels nearly as much, as long as there's no clipping upstream from the recorder.
 
Good limiters (and knowing how to hit them properly) are a great asset in the field.
Also, if you're mixing and booming at the same time there has to be compromises so I will usually play the boom farther away from them as they scream..moving it back closer
as they get quieter. All as seamless as possible of course.

Between these 2 things and having the experience to anticipate level changes BEFORE they do it, you'll be covered pretty well.
In time you'll actually be able to tell how someone is going to deliver dialog just by the way they breathe right before, or facial expression or 100 other subtle things.
 
that's a great suggestion c2v you can easily run two separate mono mixes with separate gain settings on an fp33 or any decent mixer.
 
Rippie makes a good point. With experience comes knowledge. You should see the pro union boom ops work. It's really quite impressive. I don't know how they do it. It's all a learning process, and making mistakes is part of that...not saying that the OP made any mistakes here.
 
I would say a decent limiter would be very high on my list. It's one of the things I love about the MixPre.

On most student films there is very little sound post so the "noise" in your tracks may be THE background track. The only FP33 I ever played with I found quite noisy, but that may just have been the unit. Anyway I don't remember how good the limiters are, I'm assuming it has some.
The limiters on most prosumer products are worthless so I wouldn't count on what your recording to to have decent limiters.
 
It's normal there's a difference in level about 20dB or so between someone yelling, shouting and someone whispering. It has to sound good.

What I do ask is to whisper with some voice in it... Because real whispering is mostly not understandable for someone NOT noing the lines (like every viewer).
 
For very dynamic actors I like to give 12 maybe even 18 decibels of extra cushion on a second recording channel. And ALWAYS recording at 24-bit. The options to do this (record one signal to separate channels) are one reason the expensive recorders cost more. I think some of the cheaper gear has it now but it's always a life saver to have a separate channel recording "not as hot".

For whispering lines, yeah like dre said the actors have to "speak-whisper". Real whispering is way too low and can sound sibilant when raised later in post. Not to mention the room tone being way too loud for comfort.

-6 for peaks I think is too loud. Even with a good limiter I try to stay no higher than -10dBFS. And if I DO clip the program I highly recommend buying/using is Izotope RX.
 
Last edited:
Noiz2 said:
The only FP33 I ever played with I found quite noisy, but that may just have been the unit.
Click to expand...


My experience with every FP33 that I ever used/prepped for rental was the same as your experience, but before the 302, the FP33 was the standard, and was "good" at the time.
 
davefisk said:
My experience with every FP33 that I ever used/prepped for rental was the same as your experience, but before the 302, the FP33 was the standard, and was "good" at the time.
Click to expand...

Seems like ages ago that I used an FP33 (or FP32a) on a daily basis.
Amazing how far we've come ;)

I used to own an FP42 way back. Those meters were so slow you were totally guessing what you're levels were 99% of the time.
 
You must log in or register to reply here.
Back
Top