Newb question: Cleaning up audio

[Blade Borge]

I watch deleted scenes of movies, where the audio hasn't been polished, and I always wonder what exactly did the audio post production person do to make it so much better? Could someone shed some light on this? I'm sure not all situations are the same, but generally (like in a studio) what do they do to polish the audio. I just have Adobe Premiere right now, if somebody could tell me what setting I could adjust on there to help, that would be nice too, thanks!

 

[gpforet]

Your question is like asking, what setting do I use in Premiere to make the picture look good? There is literally thousands of techniques used on audio in post. Everything from HPF, LPF, eq, compression, expansion, worldizing, gating, limiting, and the list goes on and on. Add to that most dialog is striped and sliced, background sound removed between phrases, ends feathered, and then replaced with a room tone which is designed to not only convey a sense of location, but also mask deficiencies in the dialog recording, the workflow can be immense. Then add Foley, fx, and ambience, and it's realistic to say that post audio for a 90 minute feature will consume as much as 1000 hours of post sound work.

Suffice to say that it all starts with good location dialog recordings. Five minutes spent eliminating and reducing unwanted sound sources during the original dialog recording can save literally dozens and dozens of hours in post audio cleanup.

 

[Blade Borge]

Everything from HPF, LPF, eq, compression, expansion, worldizing, gating, limiting, and the list goes on and on. Add to that most dialog is striped and sliced, background sound removed between phrases, ends feathered, and then replaced with a room tone which is designed to not only convey a sense of location, but also mask deficiencies in the dialog recording

Perfect, I know what to study more on now, thanks!

 

[pixelated]

Gpforet is absolutely correct in what he's saying. If you're asking how to clean you own audio in post . . .

Rule #1 - Get the best possible audio to start with. Check out the sound for film & television DVD in the DVXUSER shop. Excellent. It's WAY easier to get right the first time than correct later.

When you get to post - bust it all apart. It's very normal to have literally dozens of audio tracks overlaid on each frame. Each actor would have his own dialog track, at the very least filtered & equalized, a track for room tone (you did get good room tones didn't you), seperate Foley tracks, background tracks, etc each individually leveled and feathered . . .and we haven't even mentioned the score. In short, it's a huge bottomless time pit that can stretch to near infinity but can pay huge dividends if even a reasonable amount of effort is expended. Lots of things can kill a film, but bad sound is definitely the quickest.

Jay Rose's "Audio Postproduction for Digital Video" is a good first read on the subject.
http://www.amazon.com/Audio-Postproduction-Digital-Video-Rose/dp/1578201160

 

[Noiz2]

All the answers are correct to a point but don't actually answer your question exactly.

A few things. One the films you are talking about did a hollywood post workflow and not many micro budget films do that.

In a big budget film you will have dialog editors who edit dialog and nothing else. To answer you on that level, it's 90% editing. Traditionally the dialog editors didn't process (with plugins etc) the sound at all. They edited. The rerecording mixer is the one who would apply filters etc.

In lower budget films the lines get blurry and when you are the editor AND the mixer AND your doing the FX editing, then you do tend to do more "stuff" to the tracks. But it is still 90% editing.

Room tone is critical in a big budget work flow but often not so much so in low budget workflows. But a lot of people mix up "room tome" and "backgrounds" and ambiences". In a low budget film you know everything about the tracks so you don't need to "fill holes" if you know there won't be a hole to fill. In a big budget film you are working in isolation so you never know what will be out there naked for everyone to hear and look at you! So you make the dialog track work on it's own.

AS to the difference between the behind the scenes and to finished track. Part of it is all the work that has gone into it and part is that they are probably playing the crash down mix that was meant for dailies, so it may well be worse than the raw tracks. It may also be a rerecording of the dailies sound. Or even a behind the scenes camera shooting the scene with an onboard mic.

But editing would be 90% of the clean up.

ANd not counting music (which in a big picture workflow is not considered part of the "sound" team) the rule of thumb is 10 -15 man hours per screen minute, not counting the mix.

So a "normal" 90 min film will run 900 - 1350 man - hours plus say two to three weeks of pre mixes and two weeks of final mix.

"Man-hours" if you are not familiar with the term is one person working one hour. More editors, fewer days in post - but the same number of man hours.

You can do the same quality of work for around 7 man hours including the mix in a low budget workflow where you know all the facts (like where the director is going to bury the sound with a BIG score).