Changes

So, for this project the samples I used were recordings of myself reciting the text of the Bochner text as I (or another) would when interacting with the installation. The composition is broken into two types of material--actual words recited, and the pauses, gaps, and inhalations of breath between those words. My desire with the piece was to emphasize these pre- and non-verbal utterances (the physical mechanics of breath and voice) while erasing, obfuscating or overwriting the spoken text. While I think I produced some interesting material in the recordings of the silences (especially the first section of the piece) I am dissatisfied with what I accomplished my use of the spoken-word part of the text. What I would like to accomplish is a kind of redaction (similar to redacted text in classified documets) where knowledge some subterranean or obscured content is still clear, but it is impossible to decipher or extract. More on that later I suppose... Well, enough of this, it's time to listen to your pieces!  Technically, I am starting this quarter with speech-recognition and receipt printer control code implemented in Processing and java--so in a sense I have demonstrated technical feasibility. Details are in the [[#Code | Code]] section. The most readily apparent (and mundane) technical challenges for this project are the implementation of sound input and pre-processing with supercollider, and interfacing from supercollider to the speech recognition library and receipt printer. As part of this project is an investigation of the strengths and limits of speech recognition technology, I wish to delve more deeply into the mechanisms of speech recognition and the Sphinx-4 library over the course of the term. More substantial involvement with that technology is necessary to illuminate it's character and embedded values. I will update the weekly [[#Progress | Progress]] section as the quarter continues.