I’ve been working on this project I’ve been calling Dead Languages. It utilizes a custom slow scan television audio format and found objects as well as good old vhs tech. The idea is to take photos of slides I’ve found at various thrift stores, convert them to an audio file that uses a self designed encoding method, record it onto a vhs, play the vhs in an old tv and have custom software decode the result into an 8 bit image on a computer monitor alongside it.
The thing I’m dithering on is how exactly to encode the audio, I’ve had some luck decoding a custom implementation using Processing but I keep running into the same problem. I’m not very experienced with decoding real time audio using fft and I have a significant amount of data loss so the decoding is either very long or pretty glitchy. I’ve attached some images as reference of what the decoded 8 bit images look like along with the original photos
This is the github repo for it https://github.com/emmadickson/dead_languages
Some decoded images
Collage of the original slides