For midterm project, I am currently working on an interactive artwork called “Space Mood Garden”. The basic idea is to have a space‑themed screen where the user’s voice plants glowing orbs in a starry background, so their sound gradually turns into a kind of visual garden. Right now I am exploring how different levels of voice loudness can map to the size and color of these orbs, with quieter sounds creating smaller, cooler orbs and louder sounds creating larger, warmer ones. I like that the main input is the voice rather than the mouse, because it feels more personal and connects to the idea of mood, but I am still experimenting with how strong or subtle the visual changes should be.
The interaction design is starting to take shape using a three‑screen structure. The sketch begins with a start screen that shows a space background image and some simple instructions, and it waits for the user to press S and allow microphone access before moving on. In the main state, the sketch listens to the microphone using p5.sound and reads the current sound level, and whenever the volume goes above a threshold it creates a new “mood orb” at a random position on the canvas. At the moment I am mapping the volume at that moment to the orb’s size and color, and I have added a gentle pulsing animation so they look like they are breathing. 
On the coding side, I have already set up the basic structure in p5.js using a state variable to switch between the start and play.. I also created a star class for the small moving stars in the background. The microphone input is working using p5.AudioIn and getLevel when the user gives access to microphone, and I am mapping the raw volume values to the visual parameters with map. The overall system is running, but I still want to tune the timing, the cooldown between orb spawns, and the visual style so it feels more like a coherent “garden” and less like scattered dots.
For this project, the most uncertain part for me is actually how the artwork will look and feel when someone really uses it, not just the technical side of the microphone. I know in theory that the sketch will create orbs based on the user’s voice, but I am still unsure whether the final garden will feel coherent and expressive, or just like random circles scattered on a space background. Because the orbs appear at random positions and their size and color depend on sound levels that I cannot fully predict, it is hard to imagine in advance what kind of compositions people will end up with and whether those compositions will really communicate a sense of “mood”. Overall I am still actively working on the project, trying different parameter values and small design tweaks, but I feel more confident now that the core idea is solid and that with a bit more tuning the project will turn out well.