‘From the Ground Up’ – Building a Sound

You will build 4 test pieces that focus on particular areas of sound design:

  1. A.D.R
  2. SFX
  3. Foley
  4. Atmospheres

Each test piece will be an originally filmed narrative sequence, photographed by the entire group. You will then work in small groups to record, edit and mix the sound design for each. Emphasis will be placed on the art of ‘building’ a sound or a series of sounds in layers; from the ground up.


Week 4: For this assignment, we were given a scene from The mirror (1975) by Andrei Tarkovsky without any audio and our task was to build up the atmos pas (or atmospheric sounds) for it. This excluded any dialogue or foley sounds ect. Atmos pas is usually the first track in sound design and sometimes can be done in pre-production even before principle photography. It is because the sounds aren’t ‘direct’ or ‘synced’ that this is able to be done beforehand.

Atmos Pas:

For this task I worked together with Vicky Grant. The scene we were given, the character is a dream state. We  wanted the Theme to be eerie and unrealistic. The Mood we were going for was strange and unsettling. In the Narrative Arch, we felt like this was definitely building up to something so it had to be in the complication stage. Location is indoors in a creaky old house and the Weather is really windy and cold. With the water drop we played them in reverse to create a more dream like state.

Screen shot 2013-10-09 at 13.39.13

Here is a screen shot of us using the audio mixer in Adobe Premier Pro CS3. Each sound has its own track and was named properly to make things easier. We adjusted the pitch levels when we needed them higher or lower depending on what we were going for. We even tried adjusting the balance from left to right with the wind but ended not using it. Below is our final product and below that is the original.

Week 5:

For this weeks assignment our class focused on making Foley Sounds as well as Spot FX. We did a test piece from the animation Vexille. We decided what foley sounds we would need, for the humans we would need clothes rubbing together and footsteps with solid shoes on snow and concrete stairs. For the robots we would need robotic sounds of some sort but we weren’t quite sure what would work. Below is a picture of the studio, we used three microphones 2 hyper cardioid  mics for dialogue and one shotgun microphone for spot effects. After having problems with Pro Tools we decided that Apples Garageband would work nicely.


Below is a picture of the sound mixer that we had all the XLR leads running into from the microphones. They were all in their proper order running from left to right. Here we checked our levels and made sure none of our sounds peaked as we recorded.IMG_3325

Craig Ellis did a lot of our foley sounds for the humans. We found some solid shoes for him to wear and in sync he mimicked the movements of the characters on the projected screen. For the surface sounds we found a couple slabs of concrete for our hard surface and we decided on a shower curtain for our snowy surface. Both worked rather nicely. Below is a picture of Craig rubbing together pieces of cloths using the shotgun microphone to imitate the noise trousers make while walking.

For the robot sounds we used a number of items which included a squeaky desk lamp, an old dial telephone, a VHS player, some sort of dial, and a desk chair. All the sounds layered on each other worked almost too perfectly. It was a matter of trial and error and getting sounds matched up perfectly but in the end we got there. At one point we couldn’t figure out anything to use for the robots scanner, So I thought I’d give vocalizing it could work. Sadly it didn’t but it was fun to try. One thing we were unable to work on because we ran out of time was adjusting the pitch and balance to give our sounds depth. This will be something that I really look forward to doing.


Week 6:

For this weeks assignment we focused on ADR (Automated Dialogue Replacement) for the animation Vexille. We discovered that since the dialogue was originally Japanese the timing wouldn’t be dubbed exactly right. We worked around this as best as we could. We used the hyper cardioid microphones as these were best fit for what we were doing. We figured out that if the dialogue required yelling into the microphone that we could distance ourselves further from the microphone as it was louder and we wouldn’t restrict ourselves saying our lines, and our audio waveforms wouldn’t peak.


At one point we realized that the dialogue had some reverb attached to it, for some odd reason. We were unable to figure it out until Joey Lever mentioned that Garageband put some reverb on it if you select a recording because it’s mainly used for singing and it’s an added effect Garageband throws on there. If you selected a no effect real instrument it comes out nice and non-disrupted. It took us a while to get everything to sound right, but with loads of trial and error, in the end we got there.

Criteria: 2.1 & 3.1