SONIC DESIGN / ASSIGNMENT 03

Naura / 0356798 / Interactive Spatial Design
VSA60304 / Sonic Design
Assignment 03: Audio Storytelling

Instructions:

• Choose a popular fairy tale of your choice.
• Create an audiobook version of the story.
• Include:
  • Voice Acting: Record narrations with suitable voice expressions.
  • Sound Effects: Add soundscapes and effects to enhance the story.
• Combine all elements to create a complete, engaging storytelling experience.
• Imagine this as a radio show production.

__________________________________________________________

For this assignment, we were tasked with retelling a fairy tale. I chose The Princess and the Pea because it’s a story many people are familiar with. The version I worked with was retold by Shari Eskenas. When I initially recorded myself reading the full-length version, it was about five minutes long. To meet the required time limit of around three minutes, I edited the script and removed certain parts to condense it. I had considered asking a friend with a posh accent to narrate, but coordinating their schedule and managing background noise seemed too complicated.

Fig 1.0 | The Book I've Chosen

To record my narration, I tried to find a time when outside noise was minimal, as I live near the LRT and its sound can occasionally creep into recordings. I also turned off the air conditioning and fans to minimize background noise. For additional soundproofing, I wrapped myself in a blanket while recording.

My process involved recording the entire narration in one go, aiming for as much perfection as possible. If I made a mistake, I left a pause and restarted from that point. I ended up doing three takes because I wasn’t always satisfied with my timing, intonation, or my attempt at dialogue accents, which didn’t sound as I had imagined.

Fig 1.1 | My Recording Takes

In Adobe Audition, I began by removing static background noise from all three recording takes. To do this, I captured a noise print from the first five seconds of each take and applied noise reduction processing. Since I controlled the recording environment as much as possible, the settings for noise reduction were generally consistent across all takes. However, slight variations occurred due to differences in the distance between myself and the microphone during each recording.

Fig 1.2 | Noise Reduction (Process) Settings

I also applied similar adjustments using the dynamics effect rack. By tweaking the settings, I was able to enhance the audio quality and ensure consistency throughout the narration. Here are the specific settings I used for this effect:

Fig 1.3 | Dynamics Settings

I compared all three takes to identify the best segments for each part of the story. I then combined those segments into a new track and renamed each section based on the corresponding pages of the narration. To polish the audio, I manually removed lip-smacking sounds and trimmed parts where there were overly long pauses. This ensured the narration flowed naturally and sounded cohesive.

For the sound effects, I added them gradually as I worked through the project, incorporating them wherever they felt necessary to enhance the storytelling.

I adjusted some of the sound effects to better fit the story. For example, the knocking sound effect needed to reflect the setting of a castle, so I made it more echoey and heavy. To achieve this, I adjusted the EQ to emphasize lower frequencies and added reverb to create a spacious, resonant effect that matched the ambiance of a castle.

Fig 1.4 | EQ Settings for Knocking and Opening Castle Door

Fig1.5 | Reverb Settings for an Echoey and Spacious Effect

During the cleanup process, I noticed that some of my "S" sounds were quite distorted and could be harsh on the ears. To address this, I used a "De-Esser" tool to soften those sounds. I followed this tutorial (YouTube Link) to identify the specific frequencies causing the issue. However, using the spectral frequency display to pinpoint these distortions had to be done individually for each narration segment, which was a bit tedious. Ultimately, I discovered that most of the distortions occurred within the 6kHz to 12kHz frequency range.

Fig 1.6 | De-Esser Setting Adjustments

For the dialogues, I wanted to modify my voice to better fit the characters. In my interpretation, the queen had a childish voice because her decision to test whether the girl was a real princess seemed a bit silly. To achieve this, I highlighted the sections containing the queen's dialogue and created a new track. Then, I used a pitch shifter to increase the semitones slightly, making the voice sound more feminine and playful. Once I was satisfied with the result, I copied the adjusted dialogue back into the original track.

Fig 1.7 | Queen Pitch-Shifter Settings

Fig 1.8 | Princess Pitch-Shifter Settings

Consultation:

Sir mentioned that my narration sounded too echoey and offered some advice to improve the recording quality. He suggested recording inside a closet or a car for better acoustics (I don't have both) or using a thicker blanket (but as a broke international student, I don’t have the means to one). So unfortunately, none of these options are feasible for me at the moment.

He advised me to adjust the EQ on my narration since my voice sounded a bit muffled. He provided me with EQ settings to improve the audio, which I applied to my first version. If the issue persists, I might have to ask one of my siblings, to help record the narration for me.

Fig 1.9 | EQ Settings for Narration

Fig 1.10 | First Version of My Third Assignment

Second Revision:

I managed to get my older sister to record the narration—yay! It was a bit last-minute because of the 5-hour time difference, but we made it work. The story is a bit longer now though. But I processed all 13 parts of her recordings using the same noise reduction settings to maintain consistency. After that, I manually cleaned up the audio by removing parts where she took breaths, lip-smacking sounds, and reducing the intensity of her plosive sounds to ensure a smoother and more polished final result.

Fig 2.0 | Removing taking in breaths

However, for the rack effect dynamics and deEsser, the settings varied differently than mine. Below are the adjustments I made:

Fig 2.1 | Rack Effects Dynamics

Fig 2.2 | To Find the Distorted Frequencies

Fig 2.3 | DeEsser Settings

Thankfully, this process was relatively quick—haha—since I have other assignments piling up that I need to catch up on!

Fig 2.4 | Multitrack Overview of my Third ASMT

To finalize everything, I used CapCut to stitch the video and audio together seamlessly.

Fig 2.4 | Combining Video and Audio

Below is my final outcome for this assignment:

Reflection:

This assignment was definitely a rollercoaster ride—it was SO awkward to listen to my own voice! I cringed every time I had to do the dialogues, but honestly, it was fun in the end. One thing I really enjoyed was noticing how, even though my sister and I were reading the same story, our pacing and intonations were different in certain sections, adding unique dynamics to the narration.

Redoing the narration felt like a hassle at first, but I now understand the difference in quality and tone when the voice is clear versus when it sounds echoey. In hindsight, I’m really glad I decided to redo it—it made the final product much better and more polished!