Any views expressed within media held on this service are those of the contributors, should not be taken as approved or endorsed by the University, and do not necessarily reflect the views of the University in respect of any particular issue.

Category: Group 2 Audio Mixing

Big dog and small dog mixing

In the “large dog vs small dog” section of the video, since the camera perspective switches back and forth between the two dogs, I applied a similar approach in the sound design. Specifically, I alternated between the two pre-recorded and processed environmental ambience tracks, matching the shift in perspective. This allows the audience to clearly perceive the difference in spatial hearing between the two dogs.

Beyond the environmental ambience, my main focus during the sound design process was adjusting all sound elements except for the dogs’ own vocalisations, especially the EQ and tonal treatment of sound effects and human speech.

From a dog’s perspective, language is not fully comprehensible—what they pick up on are mainly tones, short commands, and key phrases. So, I used AI to generate a segment of human dialogue. I preserved the parts that sounded like clear commands or recognisable short phrases, while processing the rest to obscure the words. The result is a voice that maintains intonation and emotional tone, but becomes unintelligible, simulating how a dog might hear someone speaking without understanding the language.

Additionally, I made a clear distinction between the owner’s voice and the voices of other people in the environment. In a dog’s world, the owner’s voice holds unique emotional weight and should sound different from everyone else.

For the owner’s voice, I used a combination of Phat FX and Step FX. This blend created a sound that is partially unintelligible yet emotionally expressive, preserving the rhythm and tone without full clarity. It contrasts with the later segments where commands are delivered unprocessed, helping to distinguish the emotional impact of meaningful phrases.

 

For ambient crowd voices and general human chatter, I applied only Phat FX. This gives the sound a more distorted, less emotionally direct quality, where the language becomes vague and the tone more abstract, creating a sonic contrast to the owner’s voice.

Finally, I adjusted the EQ of all non-dog-originated sounds (environment, effects, and speech) based on the dog’s size and presumed hearing characteristics:
For larger dogs, I boosted low frequencies and reduced highs, creating a broader, fuller sense of hearing.
For smaller dogs, like Chihuahuas, I enhanced the high frequencies and cut some lows, narrowing the sound field to make it sharper and more focused.

Through all of these audio decisions, my goal was to ensure that the audience not only sees the world through each dog’s eyes but also hears the world as each dog might—highlighting how size, focus, and emotional connection shape the canine listening experience.

Ambience Sound Recording

Zixuan：

Today, we recorded environmental sound using the Sennheiser AMBEO VR microphone. We captured three separate recordings at the same location, with the only difference being the recording height, to simulate how dogs of different sizes perceive their surroundings.

Since the Chihuahua is very small, we couldn’t find a mic stand low enough to match its ear level. So we rested the microphone directly on the mic stand at a low angle to approximate its actual height.
For the Labrador and the Irish Wolfhound, we recorded at approximately 60 cm and 120 cm from the ground, respectively, to match their standing ear positions.

After recording, I processed the environmental sound recordings for the large dog and small dog perspectives, making adjustments based on their body size and likely auditory characteristics.

Larger dogs (such as the Irish Wolfhound) have larger body sizes and ear membrane areas, which make them generally more sensitive to low frequencies and less responsive to high frequencies. So, in post-processing, I boosted the low frequencies and slightly reduced the highs while also widening the stereo image to create a broader, fuller auditory space.

Smaller dogs (like the Chihuahua) are typically more sensitive to high frequencies but less responsive to lows. Therefore, I enhanced the high frequencies, reduced some of the lows, and narrowed the overall sound field to create a more focused, sharper listening perspective.

With these adjustments, we aim to authentically simulate how dogs of different sizes hear the world, enhancing the immersive quality of the experience and reinforcing the concept of “listening from a dog’s perspective.”

Ruiqi:

That was what I did on atmos for Labrador. Dogs typically hear frequencies from 40 Hz to 45,000 Hz, way far exceeding human hearing (20 Hz–20,000 Hz). They are most sensitive to higher frequencies (2k–45 kHz), which are critical for detecting sounds like prey movements, high-pitched whistles and orders. And I think amplifying 4k Hz can make sounds like human footsteps and verbal commands more perceptible.

Carly:

When recording the the ambience I made sure of having a measuring tape so we could get the correct size, the thing is that while recording the ambience it was still the plan to use the photos taken with the camera rather than the phone, which is the reason we had the measuring tape too so we could have the correct height.

Making the dog heads

Zixuan：

🐶 Dog Head Microphone – Making Process Log

Today we completed the first round of making our dog head microphone. In terms of materials and structural design, we did a bit of experimentation in order to simulate the physical shape and auditory perspective of a real dog’s head using simple materials. This will help us more accurately capture sound from a “dog’s point of hearing” for our project.

📐 Dog Head Sizes and Breed Categories

Based on the three types of dogs featured in our project, we created head models of different sizes:

– Chihuahua (small breed): The actual head diameter is about 6–8 cm. We chose 8 cm as our reference.
– Labrador Retriever (medium breed): Scaled up to a 20 cm head diameter.
– Irish Wolfhound (large breed): For the greatest contrast, we made the head about 35 cm in diameter.

By creating models with these size differences, we aim to simulate how dogs of different breeds and sizes might experience sound spatially. This gives us a physical foundation for our sound design.

🔧 Materials and Construction Approach

We used three main materials to build the dog heads:

– Cardboard: Served as the skeletal framework—lightweight, easy to cut and shape, yet strong enough for structure.
– Plasticine: Simulated the “muscle layer” and soft tissue of a dog’s head, giving weight and flexibility to the shape.
– White glue (PVA): Brushed on the outside as a sealing layer to strengthen the structure, prevent deformation, and add texture.

An unexpected bonus of this combination is the tactile and acoustic feedback—when you tap the completed dog head, it produces a sound somewhere between soft flesh and solid form, very similar to striking actual animal tissue. This quality gave us ideas for using these models not only for microphone placement but potentially as physical sound sources for future sampling.

🛠 Challenges and Solutions During Construction

During the process, we ran into a shortage of cardboard. To make the most of what we had, we carefully selected the best pieces and **patched them together using transparent tape** to form the frame.

Initially, we were concerned about whether the taped structure would hold, but once the clay was added over it, the overall form became much more solid and stable, more than we expected.

In terms of practical functionality, we also designed the heads with recording in mind. We left microphone slots on both sides of the head, aligning roughly with where a dog’s ears would be, and created openings at the bottom and side for cable routing. This setup allows us to mount mics easily and connect them to recording devices, achieving a binaural recording experience that closely mimics a dog’s hearing position.

Despite the simplicity of the materials, this handmade process gave us valuable insights into structural design, texture, and acoustic response. It also laid a solid foundation for the next phase of our work—testing how the different-sized dog heads affect recorded sound in the same environment, and further exploring the relationship between body structure and auditory space.

Ruiqi:

I hate to say this, but… I’ll never forget how embarrassing it was, running into every single shop on South Clerk Street asking for cardboard.

Carly:

It was a shock to discover that the plasticine here is different from the one in Spain. I think one of the biggest problems we encountered was how long the heads took compared to how long we actually planned them to take. We started at 10 am, thinking we would finish around 2 pm or 3 pm. We didn’t even finish that day…

Group meeting 9

Zixuan：

Today’s team meeting finalised the core direction of our project. After the discussion, we decided to focus on “a dog’s perspective” by combining video and photography to show how dogs experience the world. We chose Calton Hill as our presentation location because it offers both open natural scenery and the opportunity for the audience to gain a human perspective and find out the difference between humans and dogs.

For content design, we will explore two key comparisons:

1.How size affects perception:
We’ll compare Chihuahuas and Irish Wolfhounds, studying differences in their behaviour and hearing to understand how body size shapes their experience of the world.

2.Special sensory experiences:
We’ll focus on blind and deaf Labrador Retrievers to show how disabled dogs perceive their surroundings. We’ll observe how blind dogs rely on hearing and smell to navigate space, and how deaf dogs use vision and touch to understand their environment. This will highlight the unique ways dogs adapt when missing key senses.

To capture authentic sounds, we’ll build a “dog-head microphone” that mimics a dog’s ear structure for realistic audio recording. For special-needs dogs, we’ll also use vibration sensors to track their responses to different environments.

Additionally, we’ll film in multiple settings to compare how blind and deaf dogs behave in the same situations. This will help showcase their unique ways of adapting, making the project more engaging and impactful.

Mixing on Labrador

As we have recorded the background sounds according to different kinds of dogs’ sizes and hearing traits, coupled with sound effects that they would hear or make, we presented them in the mixing as much as possible.

For the blind Labrador part, I went with binaural audio — so the sound feels like it’s happening all around your head, helping listeners feel a bit more “dog” in the process.

Reaper Project of Deaf/ Blind Dog

For dogs who have issues with watching, an orange speaker was added to its soundtrack to boost the sensation of acuity and sensitivity.

Adding up the high frequency and granularity

And other details about processing are listed below for reference.

Amping up the distorted vibe
Complying with the dog’s hearing range, 40- 60 kHz

The deaf dog version is processed based on the blind dog’s configuration (but not binaural anymore, because it can only hear things on one side as discussed). However, I’ve panned the previous soundtrack to the left channel, while adding a pure sine sweep—two octaves above—into the right, amplifying the sensation of distraction and disorientation. (I suspect those dogs can’t perceive orientation at all.)

Initially, I applied a Doppler effect to the sine sweep in the presentation version. However, after discussions with the group and following Jule’s advice, I decided to retain the pure sine sweep.

At last, I want to explain why I was so determined to add various distortion plug-ins to these soundtracks. I was deeply moved and inspired by the film Sound of Metal (2019, Darius Marder), which powerfully portrays how individuals with hearing loss might perceive sound — metallic, distorted, even robotic — yet rendered in an unexpectedly simple and affecting way.

css.php

Report this page

To report inappropriate content on this page, please use the form below. Upon receiving your report, we will be in touch as per the Take Down Policy of the service.

Please note that personal data collected through this form is used and stored for the purposes of processing this report and communication with you.

If you are unable to report a concern about content via this form please contact the Service Owner.

Please enter an email address you wish to be contacted on. Please describe the unacceptable content in sufficient detail to allow us to locate it, and why you consider it to be unacceptable.
By submitting this report, you accept that it is accurate and that fraudulent or nuisance complaints may result in action by the University.

  Cancel