Task 3: Audio Storytelling
Task 3 requires us to choose a fairy tale and record narration and voice-over. After recording, it needs to be edited in Adobe Audition to improve the sound quality, and appropriate sound effects and music should be added. Finally, the exported audio should be dubbed along with screenshots of storybook pages or movie/TV show/illustration screenshots. The narration and voice-over can be done by a friend. It is recommended to use a mobile device for recording.
For this assignment, I chose the classic fairy tale "Rapunzel" as the subject of my adaptation and production. My goal was to create an audiobook video of about six minutes in length, which would not only include engaging narration and character dialogue, but also, through carefully designed sound effects and background music, build an immersive auditory world for the listener, making them feel as if they were witnessing the lonely tower in the forest and experiencing the characters' joys and sorrows.
Story script
Task 3 story.docx，作者 tian dong After finalizing the story script, I began the recording phase. In the voice acting, I tried my best to differentiate the different characters: the narrator used a steady and narrative tone; Rapunzel's voice was clear and gentle, with a touch of innocent naivety; the prince's voice was calm and warm; and for the witch Gosso's voice, I lowered my voice and added some breathy sounds to express her gloom and controlling nature.
Original dubbing file
After recording, I imported all the audio materials into Adobe Audition for post-processing. First, I selected a segment from each recording containing only ambient noise and obtained noise samples using "Effects > Noise Reduction/Restoration > Capture Noise Print". Then, I selected the entire vocal segment and used "Effects > Noise Reduction/Restoration > Noise Reduction (Processing)" to set the noise reduction level between 70% and 80%, carefully adjusting the threshold and frequency range to find a balance between eliminating background noise and preserving the naturalness of the vocals. Excessive noise reduction can make the sound flat and distorted, so this process requires repeated listening and fine-tuning.
After cleaning up the background noise, the next step was to make the vocals sound clearer and more professional. I noticed that some words in the original recording had sudden, high volume peaks, while other sentences were relatively soft. I first manually adjusted these overly prominent peaks, and then used the compressor in the "Dynamic Processing" effects. I set the threshold to around -20dB, the ratio to 3:1, and short start and release times. This made the overall vocal volume range more focused and sound smoother, without sudden fluctuations. Next, I used automatic gates to further process any remaining faint breaths and sibilance in the recording, setting a suitable threshold to mute sounds below that value, thus making the pauses in the vocals cleaner. Finally, I applied a hard limiter to limit the output peak to -3dB to prevent popping in the final export, and added a slight input boost to make the sound more powerful.
In selecting the background music, I mainly chose two categories: soft, slightly melancholic piano and string pieces were used during Rapunzel's moments alone and tender conversations with the prince to enhance the emotions; while in tense scenes, such as when the witch discovers the secret and pushes the prince off the stage, music with dissonant tones and low rhythms was used to create a tense and uneasy atmosphere. The volume of all music and sound effects was carefully adjusted to ensure that they enhanced the atmosphere without overshadowing or interfering with the main vocal narrative.
Finally, I exported all the mixed audio as a single WAV file and imported it into Adobe Premiere Pro for video compositing. I carefully selected and created a series of illustrations for the story, precisely aligning the audio tracks with these visual elements to make the entire video's audiovisual experience smoother and more engaging.
TASK 3 document:
Adobe Audition source files
Image
Rapunzel video
Audio files
(Week 10) After completing and submitting my initial audio storytelling piece for *Tangled*, I received valuable feedback from my teacher.
The teacher pointed out that the background music was slightly abrupt in some sections, overshadowing the clarity of the vocals; the overall soundstage lacked depth, making the dialogue between characters sound flat.
Additionally, there were subtle breathing and saliva sounds in the vocals, affecting the overall clarity of the listening experience.
Furthermore, the dynamic range of the overall audio could be further controlled to achieve a more balanced and uniform volume, and the script was slightly too long and needed to be shortened. Based on this feedback, I made a series of targeted and in-depth revisions and rework.
First, I started with the most fundamental aspect: vocal purity. In my initial processing, I primarily relied on Adobe Audition's basic noise reduction tools, but this wasn't refined enough to handle unavoidable lip and teeth sounds, subtle breathing noises, and saliva sounds that are difficult to detect during recording.
In this revision, I introduced the more professional RX 7 plugin to address these challenges. I used its sibilance removal and breathing noise control functions to meticulously process each segment and each issue of the vocal track.
While this process was time-consuming, the results were significant. It was like giving the audio a deep "cleaning," making the characters' dialogue sound exceptionally clean and clear, greatly enhancing the professional listening experience. Listeners can now fully focus on the plot and performance itself, without being distracted by recording flaws.
After resolving the basic sound quality issues, I turned my attention to mixing and creating a sense of space.
Following my teacher's advice, I significantly reduced the overall volume of the background music, ensuring it always served to support the narrative, contributing solely to emotion and atmosphere, and never competing with key dialogue.
Simultaneously, to achieve the "stereoscopic" feel my teacher desired, I completely redesigned the soundscape for the character dialogues. Instead of placing all sounds in the center of the sound field, I cleverly allocated them according to the plot and character positions:
For example, when the prince speaks to Rapunzel at the foot of the tower, his voice is slightly to the left, while Rapunzel's voice is slightly to the right, simulating spatial height and distance; when the witch appears from the side of the screen or confronts the prince, her voice is also reflected in the corresponding channel. This flexible use of the left and right channels successfully transformed a flat audio narrative into a stereoscopic sound field with width and depth, greatly enhancing the story's immersion and dramatic impact.
Next, I focused on optimizing the dynamic balance and overall loudness of the audio.
I re-examined all silent segments in the audio, ensuring there were no residual blank noises and making the pauses cleaner and crisper.
Then, I added a more refined multi-band compressor to the main output bus. Unlike the single compression I used before, multi-band compression allowed me to dynamically control the low, mid, and high frequencies separately.
I focused on controlling the mid-range frequencies where vocals are concentrated, making them more prominent and smooth, while moderately tightening the low frequencies to prevent muddiness from the drum beats or ambient sound effects in the background music.
By adjusting the thresholds, ratios, and release times of each band, I ultimately maintained a highly consistent and comfortable listenable volume level throughout the entire six-minute audio, from gentle whispers to intense emotional outbursts.
Finally, to meet the time requirements and improve the narrative pacing, I went back and refined the written script. I removed some redundant descriptive sentences, merged some dialogue rounds, and successfully compressed the total story length to just over three minutes while retaining the core plot and emotional threads.
This change forced me to reshape the narrative structure, making the story more compact and fast-paced, and using "to be continued" to leave room for interpretation and make the story more open-ended.
Story document
文稿(2).docx，作者 tian dong
After this round of comprehensive and meticulous revisions, my audio work of "Rapunzel" has been improved in both technical quality and artistic expression.
Finally, I exported all the improved and mixed audio as a single WAV file and imported it into Adobe Premiere Pro for video compositing.
Audio document-wav.
Video document - (Video with Audio.MP4.)
All documents
Reflection
In completing this Rapunzel-based audio storytelling task, I gradually realized that audio storytelling is not merely recording and splicing, but a creative approach to "creating an experience." Initially, when conceiving the script, I thought that as long as the story was clearly told, that would be enough. However, after actually starting production, I discovered that building immersion was far more complex than I had imagined. Every line of dialogue, every ambient sound, and the rhythm and volume of every piece of background music affects the listener's experience. Therefore, the first thing I learned in this process was that I must design the sound from the listener's perspective, rather than simply piling up audio tracks according to my own ideas.
Technically, from initial unfamiliarity to gradually becoming familiar with Adobe Audition, I also deeply realized that "post-production audio processing" is far more than just adjusting volume. Noise reduction, compression, mouth sound processing, stereo spatial arrangement... these seemingly professional steps directly determine the quality of the work. In my first mix, the background music drowned out the vocals, there was no sense of space between dialogues, and the rhythm was slightly dragging—problems that only became apparent after repeated listening. Later, with my teacher's guidance, I began to carefully adjust every detail, using professional plugins to remove background noise and repositioning the audio channels to give the sound "space" and "direction." Although the process was tedious, I clearly felt my progress when I heard the revised version—cleaner, clearer, and more three-dimensional.
Story pacing was also a key focus of my learning. The original script was too long, with too much descriptive language, making the overall listening experience less concise. When I tried cutting content and condensing the narrative, I was surprised to find that the story became more powerful. This process helped me understand the balance between "expression" and "pacing": sometimes, reduction isn't a loss, but rather makes it easier for the listener to enter the emotion and context. This grasp of pacing is something I didn't usually pay attention to when designing or writing reports, but it was very evident in audio creation.
The most important reflection is that I rediscovered "patience" and "attention to detail" through this project. Previously, I often thought that creating a work was simply about completing one version, but this experience made me truly understand the meaning of "repeated refinement." Every time I thought it was good enough, I would discover new problems upon listening again. I've gradually become more willing to invest time in improvement, rather than rushing to "finish" the work. This change in mindset is significant for me because it will influence my future creative habits in areas such as spatial installations and interactive design.
Looking back, this Audio Storytelling project wasn't just a sound editing exercise; it was more like a comprehensive learning experience that trained me on "how to tell stories with sound, how to create experiences with technology, and how to refine my work with attitude." I'm also starting to imagine whether incorporating sound experiences into spatial interactive installations in the future would create a stronger emotional impact. I hope to continue strengthening my technical skills and maintain this focus on experience, detail, and the audience in my creative work.
Comments
Post a Comment