I am fascinated by the way in which speech, music, and other sounds in television shows and movies are conveyed textually in closed captions. Several years ago I noticed that when I tried to take a screen shot of a show with the audio caption of a particular moment, the filmed image itself is often blocked from capture due to digital content management restrictions. This produces a plain black image with the plain white text of the caption, detached from the visual context it is supposed to accompany. I became obsessed with the effect of these free-floating sonic descriptions, and during the Covid 19 lockdown I began channeling my anxiety and compulsive phone use into the creation of alternate images for the detached closed-caption texts to accompany. Eventually this compulsion lead to the creation of hundreds of individual images, most created entirely on my phone. These images are usually created with a very simple “seed image,” perhaps a simple geometric shape or a photo of some construction materials or roadside debris, which I manually manipulate through semi-fractal processes. As the seed image replicates and transforms exponentially, it may on the symmetry of a mandala or the weave of textiles. The mismatched images and captions combine emotional timbres to suggest snippets from unseen stories. I also harbor a faint hope that they make it a little harder for web algorithms to get a bead on me.