Narrative description

A mix of fabric and paper flowers. Some fabric flowers have sensors and speakers sewn onto them, which are all connected to an Arduino. There are two sensors, a wind sensor and a light sensor. The wind sensor also has a temperature sensor built in. The current weather receives a score from 0-100 based on the temperature, wind, and light and their distances from an ideal value. Then, based on this score, a track is chosen and played. The volume is changed as well, where scores closer to 0 and 100 are louder, and at 50, they are lowest. 

Progress images

Process Reflection

What was easy: 

I knew the basics of using the sewing machine and how to hand sew. I think this made the process easier for creating all the flowers. Furthermore, once I had the pattern for the flowers, it was pretty easy to sew up multiple of them. Coding was pretty simple because there were example files for each of the sensors that had code on how to interpret the raw data. Thus, I mostly used the example code to read the data and came up with the scoring system. 

What was hard: 

There were both things that were expectedly difficult and things that were unexpectedly difficult. I knew that sewing all the flowers would take a lot of time because I was not super experienced with speeding through the sewing machine and I was a pretty slow hand-sewer. I also knew that soldering and figuring out how the sensors worked would take a little more time because I had never worked with them before. Figuring out how to make the score change based on the sensor data was unexpectedly difficult. Given what could be changed within a short amount of time (light, wind), the score values did not change all that much when I was testing. This frustrated me a lot because the math theoretically made sense, but I could not figure out why the score did not change as drastically as I thought it would. Another thing that was unexpectedly difficult was figuring out in what way I should change the volume because I felt that it would not make much sense to an audience without context with my first iteration of volume control. It was also pretty difficult wrapping my head around the idea that in the Music Maker Shield, lower numbers meant louder volume. This meant that I had to mess with the math in how to determine the score since I originally set it so that a score of 100 was perfect and a score of 0 was a catastrophe-level hurricane. Another difficulty I had was choosing the tracks for each score range. I was often unsure of what kind of vibe I wanted to portray with each score. Furthermore, it was a little difficult to visualize the conditions that each score range would have because there were three factors that could all contribute to the score. 

What did I learn: 

I learned a lot in terms of the electronic components. I have never worked with a battery pack, so it took some time to learn how to power the wind sensor since it required a separate power supply. This was also my first time working with sound. I learned about how the Arduino works with sound, as well as some sound design. I also learned how to use the Cricut, which was what I used to cut out some fabric flower patterns and paper flower patterns. I also learned a little about how the Music Maker MP3 Shield could actually make its own music using the MIDI synths, but it was a little too overwhelming and late since I had already found some fitting tracks online. I would consider using the MIDI synths for the final project because that would give me some more control over what sounds would be made using the raw sensor data instead of using a scoring system and preset tracks which brought me so much difficulty. Furthermore, using the MIDI synths may give me the ability to layer sounds instead of playing a single track at a time. 

Code

https://github.com/jennnnn-z/weathered-flowers 

Next Steps

I am thinking of using the MIDI synth features of the Music Maker shield which will provide more variability in the tracks that can be generated. I am also thinking of including more interactive sensors, such as a proximity sensor and perhaps a squeeze sensor so the audience can directly affect the sounds. Furthermore, the new sensors will be able to provide some extra sounds in the context of the WQED space since the wind, light, and temperature inputs will be quite low during the time we are showcasing. Additionally, I am thinking of using bigger speakers so that the audience does not have to be so close to hear the sounds. 

Credits

Thank you to Petra who helped me a lot with learning the sound design and the fabrication of the flowers!!! 🌸 🌺 

Credits to Kevin MacLeod (Incompetech) for the tracks ("Blippy Trance", "Sincerely", "Ultralounge", and "Almost Bliss").