Spatial Audio Design / Task 3 / Audio Storytelling

29.10.2024 - 8.12.2024 (Week 6 - Week 12)
Elysa Wee Qi En / 0355060 / Bachelor of Design (Hons) in Creative Media / Taylor's University
Spatial Audio Design
Task 3 / Audio Storytelling

LECTURES

Week 9 / Microphones

Dynamic VS Condenser Microphones

• Dynamic mics are more durable
• Dropping them normally would not damage sound quality 
• Condenser mics are more fragile
• Condenser mics also need to be powered 
• By USB and Phantom Power 
• Shotgun/ Directional microphones are condenser mics as well, but they only capture sounds from the direction you point it in 
• Normally used for outdoor purposes
• Need to be pointed directly to the sound source

Field of Audio Pickup

Omnidirectional = records sound equally all around the mic 
Cardioid = captures sounds at the front most clearly, and less so at the back
Hypercardioid = captures mostly sounds at the front
Figure Eight = captures audio from the front and back, and less at the sides 
                        - can be found in some podcast or interview recording sets
                        - or when people record duets in songs
                        - but not as common nowadays

Some microphones have multiple pickup patterns, though these are more expensive. But they provide more flexibility so are worth investing in if budget allows.

The Proximity Effect

• The closer a sound source is to a directional microphone, it responds greater to lower frequencies. 
• This effect does not affect omnidirectional mics. 
• Some mics have a bass roll-off switch to counter this, but this can also be used to your advantage.

Week 9 Tutorial 

Tip: Open the Preview Editor with the button in the top right corner for easy viewing. Then you can zoom in all you want on the top and still see the full view on the bottom.

• Remember to reduce background noise as much as possible
• Select the part of background noise you want to sample
• Use Effect --> Noise Reduction --> Capture Noise Print 
• Then select the whole audio and apply the noise reduction 
• Effect --> Noise Reduction --> Noise Reduction (process) and adjust 
• Also make sure to adjust audio levels on parts where the audio is louder than the others, for example a voice clip 
• Don't just match the size of the other waveforms, make sure to listen and compare 
• Identify and reduce all imperfections to clean up
• Breathing, lip smacking etc.
• Use the Dynamics effect to set a threshold of what audio is removed/ filtered out with AutoGate
• Audio below the threshold will be removed 
• Use it to clean up the audio (breathing noises etc.)

Fig. 1.1. Using the AutoGate feature

• Compressor in the Amplitude --> Dynamics effect will automatically reduce the audio when it goes above the set threshold
• Makes the waveform look even, essentially
• Don't have the waveform be completely square, make sure to retain the dynamics of the voice (unless it's what you want) 
• Use MakeUp under the Compressor section to add volume to the track after 'squishing' the audio where it's louder 
• Be careful that the breathing and such is not audible after increasing the overall volume of the track 
• Find a balance; adjust the threshold, the ratio
• Use the compression AFTER adjusting the audio manually just to even things out 

• DeEsser effect helps remove the 'sss' sounds caused by letters like S 
• Found in Effects --> Amplitude
• Make sure to alter the CenterFrequency and Bandwidth settings to tackle the right sounds
• After applying, check if it alters the voice quality 

Bass and Treble

• Use the Frequency Analysis to check the frequencies 
• Don't have too much bass in the voice as it becomes muddy 
• Decrease the bass with the Equaliser to make the voice sound clearer
• A good frequency balance should peak at the '2' setting on the Equaliser and start decreasing from there, before rising a little again at the '5' 
• But treat this as a guide, follow your ears 

INSTRUCTIONS

Fig. 2. Module Information Booklet

Exercise 5 (Week 10): Radio Commercial 

This exercise was given to us in class. We were given a document with some short scrips, each about a paragraph long. For each script, there were descriptions on the target audience, target length of the audio and the feeling the commercial is aiming for. With that information, we were tasked with recording the narration ourselves and add music and SFX to create a radio ad.

I started by recording the voiceover. The limit was around 30 seconds. Our lecturer told us it wasn't that strict, but I wanted to try to be as close to the time limit as possible. In the industry, actual ads likely have to follow a time limit too anyway.

Fig. 3.1. Editing the voiceover [1] (28/11/2024)

Fig. 3.2. Editing the voiceover [2] (28/11/2024)

I did lean further away from the mic at the end of the recording, which is noticeable in the audio clip. I managed to slightly fix it and make it less noticeable with some effects, and also in using music and SFX to cover it a bit, but in the future I should be more careful when recording similar audio and rerecord it if I must.

Fig. 3.3. Final voiceover waveform (28/11/2024)

After fixing the voiceover as much as I could, I moved onto choosing and adding music and sound effects. I also adjusted the volume value of my voiceover track as I was a bit louder in the beginning of the recording.

Fig. 3.4. Continuing production of radio commercial (28/11/2024)

Fig. 3.5. All tracks (28/11/2024)

I didn't end up adding that many sound effects, but since the purpose of the sound effects in this project is to emphasise the words being said, I don't think it's necessarily a loss. Sometimes less is more. 

FINAL Radio Commercial Submission 

Fig. 4. Radio Commercial Submission (28/11/2024)

Music by Samet Akkol from Pixabay

SFX Credits:

Bell13.aif by lth_stp -- https://freesound.org/s/120496/ -- License: Attribution NonCommercial 3.0

Wedding Applause Cheerfully.aif by giovannapaludetto -- https://freesound.org/s/414292/ -- License: Creative Commons 0

Wine Glass clank 2 by wasserbjorn -- https://freesound.org/s/761136/ -- License: Creative Commons 0

pencil on paper scratching writing by groupe1bts -- https://freesound.org/s/448968/ -- License: Creative Commons 0

Pencil Scratch 1 by OwlStorm -- https://freesound.org/s/320151/ -- License: Creative Commons 0

Project 3: Audio Storytelling

For this project, we had to choose a story to narrate and voice act. Then we had to find sound effects online and use them to enhance the storytelling. I also had to make the audio spatial (5.1).

I decided to do a reading of the famous story Rapunzel. 

This is the video I chose for the story and images: 

https://www.youtube.com/watch?v=7XjPbE2LUK4

I typed out the section of the story that I wanted to adapt in Microsoft Word and highlighted the dialogue, using different colours for each character. I also used coloured text to describe some sound effects I wanted to include, though this wasn't all of them.

Fig. 5.1. Story for project (4/12/2024)

The first thing I did was record the narration/ voiceovers. I separated the story into two parts to record, to make editing easier and so I wouldn't have to reread the whole story so many times. 

After recording the main narration, I moved onto cleaning up the audio.

Fig. 5.2. Fixing and adjusting the voiceovers (5/12/2024)

I also had to go back and record some singing for Rapunzel, as it is mentioned in the story. I decided to record a short line from 'When Will My Life Begin?' from the movie Tangled. I also had to record some of their laughter as a sound effect for further in the story.

Fig. 5.3. Adding a muffled effect onto the singing clip (8/12/2024)

For the music, I used some of the Original Soundtracks from the movie Tangled itself. I managed to find music that fit the vibe of the scenes, so I decided to use it.

Fig. 5.4. Adding music (8/12/2024)

I added the music quite early on, but this proved to be a bit of a mistake. As I had to break up the narration to make room for extra clips like the singing or SFX in general, the timing ended up being off and I had to readjust it later on. I should take note to leave the music to later on next time.

Fig. 5.5. Adding spatial effect (8/12/2024)

Of course after all the sound effects were added, I had to make the audio spatial. So like the last project, I adjust track values in the same way I did before to create the illusion of 'space'. I partially relied on the storybook images for this, to see where the characters were on the screen at the time the audio played.

Fig. 5.6. The adjusted track values (9/12/2024)

Fig. 5.7. All tracks (9/12/2024)

FINAL Audio Storytelling Submission

Fig. 6.1. Final Audio Storytelling Submission (9/12/2024)

SFX Credits:

Fig. 6.2. Credits for all SFX used (9/12/2024)

FEEDBACK

Week 10
General Feedback:
- Better to use a flipbook-style story instead of comics so the voice acting etc. doesn't feel lacking compared to the visuals

REFLECTIONS

Experience
This assignment was fun to do as well. Apart from sound design, I also really like the concept of voice acting, so getting to do it myself was quite fun as well. The radio commercial exercise he gave us was good practice as well, not just for this assignment but for future work and projects. Overall, both these projects taught me how to read a script compellingly and that different voices are used to appeal to different demographics and target audiences. They also gave me experience in choosing suitable music for the project.

Observations
I was taught a lesson here that it is good to find appropriate music early on, so you can have a clearer idea on what vibs you're going for, but you don't have to add it immediately. It is similar to the last project where I should finalise the timing of the main content first before messing with the music. This way it saves me time and effort to readjust stuff later. 

Findings
I actually learnt from all this that I am quite good at doing narration work. Apart from this, I realised that in some circumstances, especially in cartoon settings like this, you can take creative liberties in using sound effects. In order to make it sound like the witch slid down Rapunzel's hair, it was difficult for me to find a sound effect that conveyed it properly. So instead, I found the sound of a slide whistle and overlayed it on top of a hair-brushing sound to add that cartoonish 'slide down' feel. It helps cover up the hair-brushing sound that doesn't fit perfectly, while still adding context and fun to the sound.