Editor, PennSound and Jacket2
My HiPSTAS project grew out of the audio archival work I’ve done over the past six years at PennSound, combined with recent sample-based music-making endeavors. Specifically, it marks an evolution in my attitudes towards the processes of sound restoration and a reconsideration of the ways in which our poetry audio sausage gets made: i.e. the work that goes into transforming raw analogue artifacts into polished MP3 files.
PennSound’s archival ethos is different that some of our peers’ in that we see the value in presenting less than perfect recordings along with the pristine ones: an incomplete set, even one that cuts off in mid-line might still offer something to our listeners. The same is true of recordings of sub-optimal sound quality; indeed the medium in which most of our vintage recordings come to us (the compact cassette) wasn’t designed for archival storage and its inherent limitations are often compounded by issues with the recording process and/or storage in less than ideal conditions. When I first started at PennSound, I was somewhat aggressive about cleaning up these recordings using the built-in noise reduction programs in Audition and Audacity, but when used injudiciously, the results (including heavy gating and flanger-esque resonances on the voice) could be worse than the original. In time, I’d opt for a gentler, hands-on approach using selective EQing, filtering and compression that preserved a natural noise floor and sense of spatiality, but it wasn’t until I found myself using extraneous noises from recordings — like footsteps, coughs and drones from improperly biased tape machines — to create looped musique concrete compositions that I started to recognize that the audio medium is not just a transparent vehicle for recorded performance, but also a worthy subject for study unto itself.
Therefore, my proposed project will use ARLO to undertake a Cagean reading of the artifacts of the recording medium within the PennSound archives. Some of the specific phenomena I’d like to investigate include print-through, gaps and cut-outs, issues with signal strength (either clipping or noise floor issues), hardware-related issues, and the general spatial characteristics of the venue as captured on tape. While I’m particularly interested in celebrating the nature of these “flaws,” it’s likely that my focus might also yield an audio restoration how-to of sorts for PennSound’s student workers, and might also identify recordings within our archives that are in need of cleaning up.