Sound Design
Stranger Things Post-audio production
I don't own any of this video and is not for profit. It was one of several video clips to chose from where we had to produce all the sound and music for as part of a uni project.
All the audio is my own making with a couple of samples from pixabay and youtube.
SFX
Bones:
To create the sounds of bones being twisted and damaged I recorded the twisting and snapping of a celery. I felt it sounded fake. I pitched it down and stretched it whilst layering the sounds in varying pitches and times across 11 tracks. However I could not get it to sound realistic.
I downloaded a sound-fx sample of bones breaking from youtube to use alongside my recordings. However, they all sounded the same as my recording.
To attempt this again I decided to purchase various different fruits and veg to use in place or as well as the celery recordings.
To ensure this was not a microphone issue I set up my large condenser microphone to record the sounds alongside the zoom microphone. This would give me two recordings of the sounds and provide me the best chance of getting a sound I could use. I tested this with 1 carrot first to get the positioning of the microphone correct before continuing with the other fruits.
After recording I cut up the audio and layered them across 13 tracks overlapping in places at different pitches and stretched and grouped together. I added saturation and reverb and EQ’d each track as adding EQ to the group would not remove enough of the low-ends which were too prominent in the mix. Even after cutting low-ends further using parallel processing.
I am very disappointed in the outcome of these sounds. I feel like they lack realism. I can not figure out why, It does not sound right to me. I did however make it fit better in the mix with all the editing.
During First Blackout:
For the sparks of lights taking place in this scene I used a recording of my tap running in reverse and low in the mix along with the sound of dry leaves being crushed into the microphone. I then pitched it down and stretched it to last as long as the whole scene. During parts where there are less sparks visible and images appear I stretched the audio to the maximum so it was quieter and produced an effect almost as though you look through the sparks to see the image.
Heartbeat
In the original video there is a heart monitor machine playing, signifying a medical setting. To recreate this I used a bell sound and applied an envelope to produce a sound similar to a hospital beep. I faded in the beep sound and started with high reverb and reduced it as the sound increased in volume to give the impression of being taken into the memory. I took creative liberty and adjusted the speed of the beeps and added a constant beep at the end to signify a flatline to suggest the mother in the memory died at the end. It also worked well to have this beep continue when the scene returns from the mind back to reality. It acted like a shell-shock sound after an explosion fading out and hearing returning like in.
I also added in some low kicks I made to sound like a heartbeat for another project. I got these to roughly follow the beeps of the heart monitor. It did not have to be exactly as there was a lot going on in this scene.
Lights:
The lights flickering are a prominent feature in the start and in the 3rd scene. To create this sound I recorded myself tapping a glass with my nails, shaking some plastic objects in my hand and mostly used a small segment of one of the low end rumbles of my dryer in the background. I used that over stretched and pitched down, which sounded like flickering lights to me. It also had some distortion from manipulating the waveform too much which worked really well at the 1:25 mark after the protagonist comes out of her memory and back into the hospital room. I added reverb again to suit the room as well as reintroducing the hospital ambience of the scene.
Ambience
During the first blackout:
The next scene takes place inside the protagonist's mind so I added a lot of reverb to signify a dreamy state. I used a lot of the low-end rumble sounds for this section for both ambiance and as a sfx to provide a sense of movement. I stretched them all at different amounts and reversed a few of them and adjusted the pitch and amplitude down on some to play throughout as ambience.
I layered the others across 6 different tracks and used markers to indicate what sounds should be played at what point and at what volume.
Having the low-end rumbles peak with no image on screen and then reduce when there is, I felt added to the emotion to the scene and gave the impression of movement in between these memories as if moving through the mind.
Rooms
The video takes place in three key locations, the main being a large hospital type room with smooth walls and floor, another being a smaller office and the final being supernatural, inside a mind or another dimension.
I found ambient samples of a hospital room and a large office to use in the background to convey this, with the low-end rumbles for the ambiance in the mind.
Before the ambience was added it looked like there was something missing in the video. After it was added, these holes in frequencies were filled and gave the impression of being in that style of room.
Music
The original music had strong 80’s vibes with arpeggiated, analogue synths and gated reverb drums, a signature 80’s sound. I decided to keep to my strengths and use minimalist sounds to practise my more contemporary ambient music style, while producing music that accompanies the video without changing the overall aesthetic of events happening in the scene. This consisted of multiple kicks, bass and a synth keyboard.
ADR
Mom voice:
I recorded myself saying the words “Jane, I love you” in a high voice and then increased the pitch 1 semitone. There are a lot of low-end sounds happening during this scene. Low-ends are better heard centred so I panned the mom vocals hard right and left to help them be heard. Because this scene takes place in the mind I added lots of reverb.
I used the same recording of “I love you” after the protagonist returned to her body from her mind. This was not in the original video. However I think it added more suspense, by hearing her moms words echo and fuel her upcoming attack. I changed the volume and reverb amounts on this one and kept it panned to one side. This helped prevent it from being too prominent, giving the impression the girl was repeating her moms words in her head, as opposed to hearing her say them again.
I added another copy a few seconds later with the pitch increased and some distortions added. This takes place just as the protagonist is preparing her big attack. It was to show she was repeating her moms words in her head again, but this time angry, so it was distorted and almost creepy. Further giving the impression she is using those words as her source of strength. I think this worked well and I achieved the effect I wanted.
Screams:
For the screams I tried screaming into the microphone and pitching it up to sound female. However the result was unrealistic and I found it too distracting for the scene. Also there are bigger screams later on which I would not have been able to do.
I ended up using samples from another project taken from youtube. I was happy with the realism of these samples.
There are a lot of screams in this video and I could not use the same scream too often. I found another scream sample and used this to add some variety. If there are too few sounds that get repeated too often it becomes noticeable and breaks the immersion.
The main issue with using different scream samples is that they were different people's voices, at different pitches. This proved to be a challenge to try and blend seamlessly. Through the video it was not too bad but in the starting scene the protagonist gets dragged across the floor one way and then suddenly another. I used each of the scream samples to show this. The idea was to have the sudden change in direction mask the difference in screams. Like when a sound is made and then the source is abruptly moved during the sound, like having the wind knocked out of you.
Unfortunately the result did not go quite as planned and the difference was noticeable. I attempted to reduce the difference and blend the sounds as best as I could but fading one into the other and adjusting the pitch of each until they were of similar registers. The end result was still not as hoped, but better than the original attempt.
Man Shout:
For the man shout at the end i used a creature sound and a man shouting. To blend them in to each other I had two instances of the human shout panned hard left and right, while moving in towards the centre.
In contrast to this i had the loud creature shout and set it to fade in and had in panned slightly right and move to the centre. So when the human shout centres the similar time the monster shout does to help blend one sound to the other. The fades also cross over at similar times to help the blend further.
I was happy with this compared to the initial attempts. But i feel it could have a smoother transition. These are quite different sounds and frequencies so blending them was hard. But it was an improvement from having just the monster sound or just the human.
Other:
For the other sounds I used a recording of a friend and adjusted them to fit in withe the rest of the mix. I wasn't overly happy with these sounds. I could not have them to sound TV realistic.