(Michael Chinen)

This weekend I went back to basics and did some very tape-y music on the electro-acoustically traditional themes of traffic and water. That being said I left my usual mode of synthesis-based composition at the door: this is a memoir piece that I made in Audacity. I haven’t been making much music lately, so my goal was to have fun making something that I’ll enjoy listening to in a few hours using the recordings of rain, wind, and traffic that I’ve been collecting.

The main traffic in rain was recorded outside of “The Curb” coffee shop in Kaimuki’s Waialae ave, the street where I spent most weekends of my youth hustling to sell magic cards to buy used video games in addition to 50 cent “ice cake” at the crack seed store.

Sitting at a fancy-but-gritty coffee shop that hadn’t existed a few years ago, recording the rain, I found myself comparing the smell of the rain and the air to the other places that I’ve recorded rain and similar relaxing noise in California.

Waialae Avenue is the main street that runs through Kaimuki in Oahu, where I grew up. Though it is constantly changing, I noticed that the initial scent rain there was markedly different, and possibly unique, evoking a strong nostalgia. Looking for an explanation, I discovered the phenomenon “Petrichor”, which involves accumlation of oils and bacteria in rock and stone during dry periods that are released via aerosols caused by raindrops. I became curious to know if the geographical and historical bacteria might cause this scent to have a unique signature.

From this line of thinking, I also became interested the acoustic signature of an area. I took recordings of mostly-water based sounds, from Waialae Avenue’s rain, the drizzle from a chanterelle hunt, Big Sur’s and waves, the Yuba River’s flow, and the airport that took me between these areas, and applied basic processing to emphasize their similarities and differences.

I wanted to combine some of these recordings in a piece that meshes them together, similar to how my consciousness hopped about from the roar of California’s Big Sur coast and the drizzle of my fall chanterelle hunts in the East Bay. There is also the motion of the Yuba river and wind of Ocean Beach in there. Also, I’ve enjoyed hearing “Kaze wo Atsumete” at SLC when taking a non-non-stop trip back from HNL to SFO, so it felt appropriate to put here.

On the subject of rain and computer music, I did get to train a tensorflow-wavenet model on my crappy hardware to get a rain-like signal out. It sounds somewhat like rain, but it turns out rain is similar to white noise anyway, so I probably should have started with something that had more tones in it. Hopefully I’ll probably write a post about using wavenet soon. Maybe after I get a 1080, so I won’t need to wait many days for a model.