Leading up to and during the mixing process of our studio three project Game of Thrones Sound Replacement we used quite a few techniques to alter the audio we’d recorded as well as some of the audio we’d found elsewhere to supplement our sounds with e.g. sound effects. The following is a few examples of where we used pitch and performance correction techniques as well as the reparation of some of the audio we recorded.
The Automated Dialogue Replacement we recorded required a bit of work to fit with our scene. This became obviously evident to us upon listening to it in context i.e. with ambience, sound effects etc. The first thing we noticed about it was that a few of the words didn’t line up properly with the actor’s mouths. This took quite a bit of work to correct and you can read a more detailed account of the process in Ananda’s blog. The problem was fixed mainly using elastic audio to lengthen or shorten certain words or syllables spoken by our voice actors so they fit the moving image more accurately. Elastic audio is a very handy tool to use for this problem as you can place markers to separate a sentence (for example) and stretch certain parts/words/syllables of the sentence without affecting others. A process of clip gaining was also implemented in the fixing of this problem, just to increase the emphasis on particular syllables. We found that pitch shifting was a necessary process as well. Our voice actors (who are our classmates) are all under the age of twenty-five. The actors whose voices they were replacing in the clip are all over thirty and have deeper, more mature voices. We used Time Shift in the Pro Tools Audio Suite to pitch their voices down so they sounded a bit older and perhaps more mature.
This was imperative to our ADR being convincing and without these tools we would have been “up shit creek”. As mentioned in one of my recent blogposts, Guy Gray gave us the invaluable advice of EQing our ADR to fit the space they’re in. This was the finishing touch on our replacement dialogue, giving it the convincing sound we were after.
Processing was also necessary on a lot of our sound effects. We found that the sword sounds in particular had an undesirable high-end ring to them. We addressed this problem by the use of EQ. We found that by cutting a bit of high end off and pitch shifting the sounds down slightly, they sounded a lot more like “real” swords, or at least the “real” sword sounds you hear in movies. This same method of EQ was used on the swoosh sounds that accompanied sword swings and spear throws etc. For some reason every time you heard one of the swooshes there was a horrible high-end buzz accompanying it. We didn’t know how or why it started doing that but it was fixed by the same EQ settings as we used on the swords.
The crowd sounds were at a fantastic quality thanks to Ananda and Lisa, however, we noticed upon listening/watching the scene back a few times that perhaps the crowd needed to sound larger at the start of the clip, especially as it is the introduction to the scene which includes shots of the crowd, perhaps trying to convince the viewer of the actual size of the crowd and capacity of the venue. We addressed this issue by doubling the main crowd track for that section and pitching the second track down. This definitely makes the crowd sound larger. Another method of supplementation we employed was adding more screams to fill out the ambience during the hectic violent parts of the scene. Though one of the ambience tracks was a ‘panicked crowd’ type of track which included screams, we needed some more individual screams throughout just for padding and variance. We did this using the screams we’d recorded weeks earlier along with a few scream samples we’d acquired from elsewhere.