Wow that's pretty rough. I hope you can do something with that dialogue - sounds tough.
But that's what dialog editors and re-recording mixers do. Actually, the sound of this particular project is, for the most part, pretty good.
So how are you going to get those dialogue lines to sound correct - add a bunch of reverb?
By EQing every single line if I have to. That's the challenge of working in audio post, to take a whole pile of differing pieces and make a cohesive whole out of it.
In one scene the character was on the far side of the room, quite distant from the camera, so the audio had that "roomy" hollow sound, and when I brought the levels up (increased the gain of the clips) the noise came up considerably. So I went through all of the unused/alternate takes, found the ones from the CU shots and started replacing the dialog line by line. The most important aspect is to find lines that match the delivery/intensity of the original performance. My biggest challenge was that there was one phrase ("he seemed") in the original performance that was not to be found in the alternate takes, so I found a "he" from another scene and actually had to construct the word "seemed". I found an "S" from one word, a long "E", an "M" and a "D" from other words and built the word. I had to time stretch the "E" a little and raise the pitch on the "M". I then used Vocaling to align it properly and "seemed" is now a seamless part of his dialog.
I'll give you a quick look into how I work, which a far cry from how the big dogs do it, and probably very different than the work flow of other "one-man-band" audio post folks.
My first chore is to work through the production sound. On the initial run-through I do the editing and dialog replacement. I make notes on which scenes need ADR (which doesn't happen about 80% of the time). The second chore is to do all of the ambiences. That way I can find out what dialog problems can be "hidden" in the ambiences. Then the second pass through the production sound do noise reduction and EQ matching, use a downward expander to reduce the "roomy" sound, and set up a few preliminary reverbs. I then pull in the rough mixes of the ambiences (some will be 10 or more tracks) and do the third dialog pass.
That's when I can get to work on the "fun" stuff - sound FX and Foley. I usually Foley footsteps character by character rather than scene by scene; it takes me a while to get into the flow of the character and I don't want to lose it once I've found it. This runs contrary to how the pros do it, but it works for me, and I'm not constantly changing shoes. Once all of the footsteps are done I'll do a contact and cloth pass. Contacts are handshakes, back slaps, kisses, etc., and cloth is just what it sounds like - shirts, pants, dresses, jackets and the like. (Fight scenes I'll do separately.) I then do item Foley - bottles, dishes and silverware, and other misc items used by the characters. Sound FX is just what it sounds like - doors, gun shots, vehicles, etc.
By this time the initial score is completed and I start the final construction and prepare for the mix.
That's pretty much it. I'm half way through my second pass on the dialog and have put in a little over 200 hours so far.
Anyway, that's what I do. My process is constantly being altered as I find more efficient and/or creative ways to work.
Peace,
Bob
