Mingunk – Assignment 3: Don’t Mess with Me

To start this assignment, I searched the definition of an acoustic space. From Wikipedia, an acoustic space is an acoustic environment in which sound can be heard by an observer. Based on this notion, I wanted to think outside of the box to find unique sound environments that are reflective and diffractive. Luckily, inspirations for ideas were brought to my mind as I was casually conversing with my friends. I decided to test impulse responses recorded from underwater and in my body.

To obtain a response from underwater, I had to consider an ideal location for collecting it. At first, I thought of the school’s swimming pool as the only option. However, I realized that swimming pools may have noise from in and outside of the pool. Hence, I decided to use my bathtub at home to replicate an underwater environment. By using a GoPro Hero 5 from the Hunt Library, I tested various impulses responses underwater. Few of those options were popping a balloon, splashing one of the sides of my hand, and splashing the palm. Multiple trials led me to conclude that splashing one the sides of my hand brings about the best impulse response underwater.

In order to gather a response from my body, I came to realize that my mouth is both a spacious and an accessible space. I put a recorder in my mouth after a breathe in to make my mouth more spacious. Then, I inserted a small roll of tissue in one of my nostrils to block any sounds from dispersing and left the other nostril open for my clap signal to go in through it. By allowing the input signal to transfer over to my mouth (acoustic space) through one of the nostrils, the recorder was able to collect the impulse response of a clap. I also considered popping a balloon, but the response was insignificant for applying convolution.

For one of the experimental sound convolutions, I wanted to test my comical senses further. This led to try convolving an impulse fart sound to my original signal. I looked on YouTube for various fart sounds and after multiple testings, I figured that a short and clean fart produced better convolution than dirtier farts. This allowed the original signal’s beats to be replaced by farts. As for the next experimental sound, I decided to make a more serious outcome. Due to my love for rain drop sounds, I acquired a sample from YouTube and processed it in Audacity. I used the normalization effect to make the drops more audible and the phaser effect to give it a sci-fi mood. As I was processing, I hoped this impulse response can make a rap song more dreamy and instrumental.

I chose “Bad Guy” by Billie Eilish for its hard drops and beats to convolve with the first three impulse responses. Whereas for the last response, I selected “Congratulations” by Post Malone due to its party-style rap. Since both signals had a lot of bass, I ran a high pass filter in Audacity so that gain would not exceed and produce noise when I convolve with the impulse responses.

Even though testing my mouth as an acoustic space and using a fart sound for an impulse response may sound stupid, I am very satisfied with the final outcome. Because of the fact that those not only make the listeners laugh, but also allowed me to enjoy going through the convolution processes. For the other two impulse responses, I was glad to hear the result matching what I had expected before convolving.

The size of sound files were too large to be embedded on this post so the final output sounds can accessed in the Google Drive. Link to this assignment’s Google Drive: https://drive.google.com/drive/u/1/folders/17hUPip3c3mFj_vhtDdBA8zJTC0KVkWea