The Bed of Sound Theory

S

sbryan

Well-known member
I read about the transition between cutting on matched-action to cutting on a bed of sound. This technique, as I recall, was developed by Robert Altman. I guess I just want to make sure I can apply it to what I do.

Here's the scene: an expensive restaurant.

Would I:

1) Record live audio of patrons chattering (barely audible dialogue)

2) Record audio of a violinist performing

3) Record the conversation between the main couple.

Audio Track 3 would be the master audio track, with elements of tracks 1 and 2 mixed in?

How would I do a take where dialogue connects two shots, or rather, how is that done in Hollywod?
 
Well, assuming you are using the production track, and not ADR, then the dialogue will be on it's own track. I am also assuming the violin is recorded sperately, which would also be on it's on track. The chatter of the people in the retaurant (known as walla) should either be recorded in stereo, or I bet you could just find a nice stereo recording or restaurant walla in a sound FX library. That would be in the tracks with the BG's.

Basically, your sound editorial will be broken down into a couple of catagories. Dialogue, Backgrounds, Effects, Foley, Sound Design, and Music. On a typical big budget feature, each one of those catagories will be handled by one or more people, making for a several person team. On lower budget stuff, you tend to get one or two people managing it all. So, for the big budget stuff, that mean there would be a Pro Tools session for just the Dialogue, another for BG's, another for FX...and so on. Each one of those catagories will then be pre-mixed (known as pre-dubs) by themselves, and then all of them will be brought together for a final mix. The scale of the project will determine the length of the mix. Charlie and the Chocolate Factory, for example, took 6 months just to mix alone.

So, anyway, the short answer is to get someone who has done this before to do your post sound for you. That way, they can show you what's going on and what they are doing. Then you can get a good understanding of what's going on. Just make sure you get someone with some experience (if you have a budget for it).

-Dave
 
The "normal" Hollywood way would be to silence everything but your main dialog and add everything else in post. But you mentioned Altman. He usually records LOTS of tracks. Multi lavs plants etc. If you want to do that kind of thing you need to record a pile of discrete tracks as in a lav going to it's own track for everyone who speaks, You would still probably want to minimize all the non dialog to make editing possible.
And Dave is very on point, you definitely want to get someone onboard that has some experience with this kind of thing.

SK
 
the violin preforming is usially pb , and if some one need to hear it you can give them ifbs , the back ground dialog is in post ( people speak like fish) later you can ask them to mumble during the room tone session
 
don't forget to slap a slice of verb on the background noises including violin, will sound mucho realistico (that's spanish in case you didn't know). =D
 
dave, you mentioned sound design as a separate element... id always thought it was included as, simply speaking, the E in D/M/E - could you explain that? id like to hear your opinion..

tim
 
You must log in or register to reply here.
Back
Top