Sonic Design: Task 1

4.4.2022 - 25.4.2022 (Week 1 - Week 4)
Chung Yi Ki / 0345014 / BDCM
Sonic Design
Task 1: Audio Fundamentals

---

Quick links

Exercise 1

Exercise 2

Exercise 3

Exercise 4

Exercise 5

Lecture

Week 1 / Module briefing 

For week 1, we are given an introduction to the module and expectations on the assignments throughout the semester. We are also advised to prepare a good quality headphones for the projects and was given an introduction into a sound recording room in campus and the equipment inside it. We learnt that soft materials absorb sound and hard surface objects reflect sound.

Week 2 / Nature of sound

Sound is the vibration of air molecules that vibrates the eardrum. The vibrating air molecules is called sound waves.

Fig 1.1 Sound waves travelling through the ear
Source: https://www.soundproofingcompany.com/soundproofing_101/what-is-sound

A bit of biology, sound waves would travel through our ear canal and then reach our eardrums, which would vibrate it and the cochlea, the signal will be transmitted to our brain and will be interpreted as different sound.

Fig 1.2 Diagrams of particles in different medium
Source: https://www.solpass.org/science4-5/sound/sound-standards.html?section=study-8

Sound travels quicker in solid, slower in liquid, and slowest in gas due to the difference of distances between particles. 

The study of human perception of sound is called psychoacoustics. This subject involve the perception of pitch, loudness, volume and timbre. It studies how different people perceive sound differently, e.g: how some people can study under loud music. 

Fig 1.3 Amplitude and frequency
Source: https://www.quora.com/What-is-the-amplitude-of-sound-waves

The higher the amplitude of a sound wave, the louder the sound is. The closer each wave is to each other, the higher the frequency. 

Properties of sound 

1. Pitch (frequency of sound wave, the higher the frequency, the higher the pitch)
2. Loudness
3. Timbre
4. Perceived duration (how speed of sound is perceived)
5. Envelope (structure of particular sound, when does the sound get softer or louder etc.)
6. spatialization (placement of sound in an environment) 

Fig 1.4 Hearing and voicing frequency range of different animals
Source: https://theory.labster.com/hearing-range-dbs/

Human can only hear frequency range of 20Hz to 20kHz.

Week 3 / Sound design tools 

Digital audio workstation are any sound editing software and all of them have a set of common tools that are useful for sound design. 

The five most common tools and techniques used in sound design are 

1. Layering
2. Time stretching
3. Pitch shifting
4. Reversing
5. Mouth it

Layering

Layering is the technique of placing two or more sounds on top of each other to blend and mix various sound into a new unique sound. 

Time stretching/compressing

Time stretching is the action of slowing down a sound and time compressing is the action of making a sound quicker. Both shouldn't change the pitch of the original sound. When stretching a sound too much it might sound alien. The rule of thumb is to not stretch is over 50% if you want to keep the sound and not change it to sound like other things.

Pitch shifting

Pitch shifting is the action of changing the pitch of a sound without changing its length. For example, high pitch would sound like a chipmunk voice and usually they sound fast, while a lower pitch would sound like zombies or monsters and usually they sound slower. Higher pitch also usually coreleates with a tiny being while a lower pitch coreleates with a bigger being. 

Reversing

Reversing an audio is simply reversing it. It would create weird and unnatural sound and would require layering. 

Mouth it 

Mouth it is making a sound using your mouth. This is done when nothing else can be done to get the sound you want. But it should be the last resort and it's best to avoid it.

When doing audio processing there's two types of editing: destructive and non-destructive. Destructive editing means whatever changes you make, you won't be able to undo it again once you exit the software. Non-destructive editing is the opposite. A way to do non-destructive editing in Adobe Audition is to select the file and right click and copy to new so that we aren't editing on the original sound file. 

In the reverb settings in Adobe Audition, there's two parameters called "dry" and "wet". They work similar to how opacity works in Photoshop. Increasing the dry level will lower the effect and increasing the wet level will make the effect more prominent. 

Week 4 / Volume and panning automation

Sustain - How long a sound stays loud

Decay - How long a sound goes silent

Normally the sound we hear from YouTube, videos etc., the bit depth is 16 bits, so when producing sound it's better to increase the bit depth, which is 24 bits or more. 

Playing with volume generally produces a sense of distance to the sound. Loud volume usually equals to nearer distance. It will also give you different perception of sound from its original source.

Automation

Fig 2.1 The lines on the audio clip represent clip automation,
while the lines below the audio clip in a grey area is track automation
Source: https://helpx.adobe.com/audition/using/automating-mixes-envelopes1.html

Clip automation - effects that affect on the clip only and will stay on the clip no matter where the clip is arranged

Track automation - effects that affect a specific time at a track in the multitrack window

Panning

Panning an audio means controlling the stereo balance of an audio, whether it sounds louder at the left speaker output or the right speaker output.

To have better stereo imaging, utilize stereo balance on different parts of a sound. Keep in mind the environment that we need to do sound design to, if there's people is on your left or right, then the audio should reflect that too, when you do things on your own then the sound should be centered. 

When simulating distance in voice clips, adjusting just the volume isn't enough as the crisp and clear voice still gives the sense that the voice is coming from the front of us. So, equalizer would need to be adjusted too to make the sound more muffled. Equalizer settings can also be included in track automation by clicking show envelopes in Adobe Audition and choose the specific Eq band width.  

---

Tasks

Exercise 1: Equalizing sound

For this exercise, we are given a flat sound file and 4 equalized sound file to test our headphones. Our task is to re-equalize the equalized sound file to sound as similar as possible to the flat sound file. 

Fig 1.1 Given sound files

Equalizing process

Fig 2.1 sound file eq-1 equalizer window

Fig 2.2 sound file eq-2 equalizer window

Fig 2.3 sound file eq-3 equalizer window

Fig 2.4 sound file eq-4 equalizer window

For each eq sound file, I listened and compared them together with the flat sound track in Adobe Audition using the solo function, and adjusted the curve in the equalizer accordingly. For eq-1, the bass was too low so it needed to be higher; for eq-2, the treble was too low so I adjusted it to be higher; for eq-3, the mid-range frequency needed to be higher and for eq-4 the bass needed to be lower. I also adjusted the size of the wave a bit to get a more accurate edit of them. 

Final re-equalized sound file

Fig 3.1 Final re-equalized sound files

Exercise 2: Sound shaping

For exercise 2, we are tasked to edit the given sound files using the paramatic equalizer to match certain situations. 

Fig 4.1 Given sound files

sample_voice file edit - Telephone call, cupboard voice, muffled voice

We first started with editing the sample_voice file to make it sound like a telephone call, a voice from a cupboard and an even more muffled voice. 

Fig 5.1 Telephone effect equalizer graph

Fig 5.2 Cupboard muffled effect equalizer graph

Fig 5.3 Even more muffled effect equalizer graph

For the telephone effect, I increased the mid-range and decreased the bass to get that isolated kind of static effect, and decreased the treble a bit to make the voice less sharper. For the cupboard effect, I decreased the treble to make it a lot less sharper and increased the mid range to give it a bit more nasal sound, then I increased the bass to give that muffled sound. As for the even more muffled effect, I increase the bass way more and then decrease the treble all the way down. 

Fig 5.4 First attempt at telephone effect

Fig 5.5 Further edited equalizer for telephone effect

After receiving feedbacks from Mr. Razif, I edited the equalizer for the telephone effect again to decrease more treble to make the sound less sharper, and also adjusted the frequency for node 3 and 4 to increase more mid-range. 

Fig 5.6 Walkie talkie effect equalizer graph

I then used the telephone effect eq graph and adjusted it to make it sound like a walkie talkie by increasing the volume to create distortion, increasing the mid range again and decreasing the bass. 

sample_voice edit - stadium, bathroom, airport announcement 

After that, we are tasked to edit the sound file again to make it sound like it's coming from a stadium, bathroom and from the airport announcement system. 

Fig 6.1 Stadium and bathroom effect equalizer and reverb setting

For both stadium and bathroom effect, I used the same equalizer setting to create a muffled sound. For stadium, I added more echo to the sound by mainly focusing on adjusting the decay time, diffusion, dimension and early reflection sliders. For the bathroom effect, I wanted to get a more clearer sound with a bit of echo, so I mainly focused on adjusting the early reflections, diffusion, perception and dry level.

Fig 6.2 Malaysia airlines airport announcement video

Source: https://www.youtube.com/watch?v=uYdtvMS2fFU

Fig 6.3 Airport effect equalizer and reverb settings

For the airport sound, I referenced an airport announcement video to listen how the voice sounds like. In the equalizer, I increased the mid range frequency and decreased the low and high to create a telephone-like sound. Then, in the reverb settings, I mainly focused on adjusting the decay time, dry level, diffusion, dimension and early reflection to create the clear but echoey sound coming from the announcement speaker.

Fig 6.4 First attempt at bathroom voice effect

Fig 6.5 Second attempt at bathroom effect reverb settings 

After receiving feedbacks from Mr. Razif, I increased the decay time to create more echo in the voice. 

Final sound shaping for sample_voice

Fig 6.1 Final edited sound files

Exercise 3: Sound layering and effects

For this exercise, we are tasked to edit a given explosion sound file and punch sound file with various effects in Adobe Audition and layer them. For the explosion sound, we are to elevate it by editing it with different audio effects as well as equalizer and layer them, while for the punch sound, we need to make it sound like 3 consecutive punches where it gets more and more impactful. The punch sound should be done with different effects and layering as well.

Explosion sound 

Fig 7.1 Given explosion sound

Fig 7.2 Effects used to alter the explosion sound file

Fig 7.3 Multitrack window of audio tracks layout

Fig 7.4 Equalizer settings of multitrack tracks

For the explosion sound, I wanted to make it more impactful and have some anticipation before the explosion starts. I copied the original explosion sound track to make different edited versions of it to layer it later as shown in Fig 7.3. I first used a chorus effect to distort the explosion to have it sound like multiple abrupt clear echoes, and then reversed it to create a intense anticipating sci-fi like sound. Then for the other audio tracks, I changed the pitch of the audio by using the pitch shifter and/or equalizer, I also time stretched one to make the explosion sound last longer instead of ending abruptly. Hard limiter is used where necessary to avoid clipping.

Fig 7.5 Full reverb effect on final explosion audio layering

After layering the different edited explosion audio, I used a full reverb effect in the multitrack mixer to give the final audio some echoes and make it sound more believable. 

Multiple punches sound

Fig 8.1 Given punch sound

Fig 8.2 Audio effects applied on each track in the multitrack window

Fig 8.3 Equalizer window thumbnail for teach track in multitrack window

Fig 8.4 Effects used on creating the audio in track 6,2 and 4 in multitrack

For the punch sound, as like with the explosion sound, I created copies of the original track and edited in each of the copied track. I used the chorus, reverb and stretch and pitch effect on one of the copied track to create a sound with multiple impacts happening at once with some echoey feeling. Then for the other tracks, I used pitch shifter to pitch down one of the track and for the others, I changed the audio sound to have either more or less bass or treble by using the equalizer in the multitrack window. 

Fig 8.5 Echo effect settings for track 5 in multitrack 

I also used an echo effect for the audio track in track 5 of the multitrack window. Because the original punch sound file isn't long enough for the echo to finish decaying, I generated 4 seconds of silence after the punch by going to edit > insert > silence to increase the length of the sound track. Then in the multitrack, I added a pitch shifter effect to make its pitch a bit higher and a reverb to have a bit more echoey atmosphere. 

After that, I layer the tracks to make the final audio sound like a series of fast punches where the final punch lands with echoes. The audio track with echoes was faded out at the end to avoid the echo trailing on for too long. Hard limiter was used where necessary to avoid clipping, and in the multitrack mixer too.

Final layered and edited explosion and punches

Fig 9.1 Final explosion and and punches edited audio

Exercise 4: Environment soundscape

For this exercise, we are tasked to create a ambience sound mix based on two given concept art images. We can search for suitable royalty free sound effects online and layer and edit them in Adobe Audition. 

First environment soundscape

Fig 10.1 Given Environment #1 concept art image

For Environment #1, I wanted to create a soundscape where it highlights the tree in the chamber with probably some sound akin to nature or fantasy, but still keeping the sci-fi theme as the major aspect of the soundscape. Because there's computer all around and soldiers in the image, to make it less monotonous, I wanted to create a kind of narrative where the soldiers would walk around typing on the computers to activate or close some kind of tube or chamber. 

Fig 10.2 Source links of sound effects used in Environment #1

Fig 10.3 Screen recording of Adobe Audition multitrack timeline for Environment #1

After I went to Freesound.org to search for the suitable sound effects needed to create the soundscape, I layer them and edited them where necessary to create a cohesive mix. I started with building the ambience of the center tree chamber and developed on from there. To get the nature/fantasy feeling in the sci-fi environment, I included very subtle wind chimes sound throughout the mix to make it sound magical and fantastical, similar to the music heard in soundtracks of fantasy shows/films. Leaf rustling noises are also added to give sound to the tree, but it is timed to get loud whenever an air pressure release sound is played, and them slowly gets soft again. The air pressure release sound is to give sound to the pipes. All the low humming sounds are to simulate the sound of the tree chamber, and together with the computer beeping sound, they enforce the sci-fi feeling in the soundscape. 

Fig 10.4 Pitch shifter settings of walking sound

Fig 10.5 Pitch shifter settings of laboratory bubbling ambient sound

Panning is used throughout the mix to reduce monotony and give a feeling of distance, such as when the soldiers are walking and activating things on the computer. I used a chain rattling sound effect together with the walking sound to simulate the armour and weapon of the soldiers. Pitch shifter is used where necessary to edit the sound in order to make the mix more cohesive, such as lowering the pitch of the walking audio to make it sound heavier (Fig 10.4) and lowering the pitch of the laboratory bubbling audio to make it sound less cartoony (Fig 10.5). Equalizers are used in a way to either muffle a sound more or increase its bass and/or treble for the heavy or crisp sound. Track automation of equalizer is used for the picking up weapon sound effect to give it more variation, by increasing the treble gain from low to high. 

Fig 10.6 Hard limiter settings for Environment #1 soundscape

Fig 10.7 Full reverb settings for Environment #1 soundscape

When layering the sounds, I added a full reverb in the master mix so that all tracks have a reverb effect. And after layering the sounds, I added a hard limiter in the master mix to avoid the sound mix from clipping. Hard limiter is also used in some of the tracks to boost the bass, but when increasing the bass isn't enough, like for the walking sound effect since the original audio volume is low, amplify is added to the track to increase the volume beyond the 15dB limit in the track mixer. 

Second environment soundscape

Fig 11.1 Given Environment #2 concept art image

For environment #2, I have a clearer idea on how the soundscape should be. I wanted to create a narrative where a person comes in to the factory and activates the laser one by one. Since there's two lasers in the concept art, two lasers will activate with a bit of space between the start of each other's audio. The overall feeling would be mechanical as there's big machineries, echoey and sparse water drippings since the floor is wet in the concept art. 

Fig 11.2 Source links of sound effects used in Environment #2

Fig 11.3 Screen recording of Adobe Audition multitrack timeline for Environment #2

Same with Environment #1, I downloaded all the sound effects needed from Freesound.org. Pitch shifters and equalizers were used for the same reason in Environment #1, to make the audio more suitable for the soundscape in Environment #2 and to boost bass and/or treble and muffle some audio. I used clip automation more here because I experimented with moving the clips around a lot during the layering process, so it's easier to just keep the automation adjustment with the clips.

Fig 11.4 Chorus effect settings for laser startup audio edit

Fig 11.3 Echo effect settings for laser firing audio edit

Though for Environment #2, I edited the downloaded audios more compared to the audios in Environment #1. To create the sound of the laser starting up, I used a synthesized moving machine audio and added chorus effect to give it that choppy repeating sci-fi sound. Then, I reversed the audio so that the chorus effect starts first, giving it a machinery starting up feeling. Another version of the laser startup audio is saved without reversing the audio so that both of them can be layered to represent the two lasers in the concept art without being monotonous. Other than that, I also added an echo effect on the laser firing sound effect so that the sound would trial on longer after it has been fired, giving it a huge and echoey feeling. 

Fig 11.4 Toch screen beeping and its original source audio comparison

For the beeping sound when the person is operating on the holographic interfaces, I couldn't find the exact sound that sounds right so, I downloaded a synthesized beeping sound of a crystal clock generator and edited the pitch of each beep to better simulate the variety of keying something on a computer, and manually adjusted the space between beeps so that there's a variety of fast and slow pauses. This was all done by selecting a specific area in the Adobe Audition audio waveform window and apply the needed pitch shift effect or delete excess silence. 

Fig 11.5 Full reverb settings for Environment #2 soundscape

Fig 11.6 Hard limiter settings for Environment #2 soundscape

As like Environment #1, full reverb effect was added to the master mix of the Environment #2 sound mix to give every track the echoey reverb effect. The reverb settings here produces a more echoey effect compared to that of Environment #1. Hard limiter was later used after all the audios were layered to avoid audio clipping. 

Fig 11.7 First attempt at Environment #1 and Environment #2 soundscape

Further refinement of soundscapes

Fig 12.1 Increasing the pitch and adjusting the volume for the deep bubbling sound in Environment #1

After receiving feedbacks from Mr. Razif, I went and did some tweaking on some sounds in the soundscapes. For Environment #1, I increased the pitch of the deep bubbling sound effect by 2 semitones and also increased the volume by adjusting the track automation keyframes and by applying amplify to further add 3 more decibels. This is so that the bubbling sound would be more clearer but still retaining its deep, growling tone to match the massiveness of the chamber in the center.

Fig 12.1 Addition of another laser firing sound at track 15,16,17 for Environment #2

As for Environment #2, I added a third laser firing sound by duplicating the more silent laser sounds in the first attempt of the soundscape mix, and rearrange them to start sooner than the other 2 lasers. I also adjusted the volumes for the duplicated sounds so that it sounds further away, and rearranged the other sounds so that the sequence is not compromised. 

Final environment soundscape

Fig 13.1 Final Environment #1 and Environment #2 soundscape

Exercise 5: Radio commercial

For this exercise, we are divided intro groups and are tasked to choose and record a script from a given document and enhance the recording by removing noise, clicks, and unnecessary loudness, as well as adding suitable music and/or sound effects. One person from the group will be the voice of the script while all of us will edit the recording. 

Fig 14.1 Recording of script

Fig 14.2 Photo of my group in the recording studio, taken by Mr. Razif

My group consisted of Adina, Yi Yun, Akari and me. We chose a wedding event planning commercial script and spun the wheel to see who will voice act it. Fate decided on Adlina and she agreed to record the script. We recorded the script in the campus's audio recording studio, me, Yi Yun and Akari was handling the ProTools software and recording equipment under the guidance of Mr. Razif.

Audio editing

Fig 15.1 Reducing noise in the recording

Fig 15.2 Amplifying a portion of the audio that got too soft after the noise reduction

Fig 15.3 Removing breathing and lip smacking sound by selecting and lowering the volume

Fig 15.4 Autogate settings

In the editing process, I first started with removing any noise in the recording by using a silent section for the noise print and use the noise reduction (process) effect. Then I search for any clicks, pops and breathing sounds and remove them by selecting and lowering their volume. I also used the compressor settings in the dynamics effect to make the volume of each section consistent with one another, and added some autogate effect as a touch up in removing any subtle clicks and pops. 

Layering with music and sound effects

Fig 16.1 Multitrack window

Fig 16.2 Version 1 of the audio layering

Fig 16.3 Version 2 of the audio layering

After finishing with cleaning up the recording, I went to search for suitable music and sound effects. I searched for the sound effects based on what was written in the script and chose two options for the music, one is the classical wedding march music while the other is more sentimental piano music. 

Music and sound effects sources:

1. A Midsummer Night's Dream, Op. 61 - Wedding March by Mendelssohn (used in version 1) 
   https://musopen.org/music/317-a-midsummer-nights-dream-op-61/ 
2. Magic Wedding by ArtIss22 (used in version 2)
   https://pixabay.com/music/wedding-magic-wedding-120021/
3. WEDDING-BELLS by Maurice_J_K (used in version 2)
   https://freesound.org/people/Maurice_J_K/sounds/346941/
4. wedding glasses.wav by shawshank73
   https://freesound.org/people/shawshank73/sounds/127252/
5. Champagne cork pop by Ruben_Uitenweerde
   https://freesound.org/people/Ruben_Uitenweerde/sounds/486161/
6. Wedding Applause Cheerfully.aif by giovannapaludetto
   https://freesound.org/people/giovannapaludetto/sounds/414292/
7. Royal sparkle whoosh centre.wav by alexkandrell
   https://freesound.org/people/alexkandrell/sounds/170523/

After comparing the two versions, I think version 1 works better since the music is able to match the giddiness and upbeat vibes of the script recording. So, I chose that to be as my final. 

Fig 16.4 Paramatic equalizer settings for Wedding March music

Fig 16.5 Denoiser settings for Wedding March music

When editing for version 1, I had to edit the Wedding March music as well because there was some noise in the music file. Since the music file doesn't have a pause long enough for me to get a noise print, I used the denoiser effect instead to decrease the noise without altering the audio, and then used the paramatic equalizer to decrease the muddiness of the sound and make it sound crispier. 

Final Radio Commercial

Fig 17.1 Final wedding event planning radio commercial

---

Feedbacks

Week 2 - Exercise 2

The telephone effect should have less treble since telephone voices aren't that crispy and clear. The bathroom effect should have more decay to give it a bit more echo.

Week 3 - Exercise 3 

The bass sounds prominent on my headphones for the edited explosion audio.

Week 5 - Exercise 4

For Environment #1, the inclusion of reverb is nice, it gives the ambient feeling that it's in a huge room. The subtle footsteps and the subtle computer sound also sounds nice. Though, if the bubbling sound is too slow, the volume might need to be louder to make the centerpiece more prominent. 

For Environment #2, the way the sound arranged is good, as it doesn't sound chaotic and sound like it has a narrative. The inclusion of reverb is good, makes the soundscape feels like it's in a big room. The water dripping sound effect is good because it fits the ambience of the picture. The laser beam sound is nice. Try and add 3 laser firing sound there with them firing alternatingly. 

---

Reflection

These exercises were really useful in getting me more familiar with the tools and layout in Adobe Audition. Although I did use Adobe Audition before back in Video and Sound Production module in Sem 2 and did the same exercise 2 and 3, doing them again did help me in refreshing my memory on using Adobe Audition since it's been a while after that module, and I mostly audio edit with video in Adobe Premiere Pro. Exercise 4 for me was the most fun and interesting one since it's my first time creating an environmental soundscape by basing on just an image, so I got to learn a lot on how to use the Multitrack and audio effects when doing those exercise. Plus, it's also a great introduction to Project 1.