ARLO

The noise is the content: Toward computationally determining the provenance of poetry recordings

Chris Mustazza

[epiphone] 

There seems to me no better way to begin this discussion than with an epistemological thought experiment (as is the case with most discussions). Consider what you heard in the “epiphone” to this essay[1], which is hiss from a digitization of recordings of Vachel Lindsay, originally made on aluminum records in 1931. It likely sounded like noise, and it is—to human auditory perception. But what if there is a pattern in this noise that is imperceptible to the human ear but recognizable to so-called machine listening? Consider the sample above from the Lindsay, alongside this sample of leading “noise” from digitizations of Harriet Monroe from the same series, alongside this one from the James Weldon Johnson recordings. I’ve been listening to several hours of audio from this series and have come to think that the noise from each of the recordings sounds similar, in the most impressionistic way possible.

Physio-digital responses to the digital

Stephen Vincent's haptic drawings of the sound of poets reading

Stephen Vincent, "Haptic: CA Conrad Reading at Nonsite," September 12, 2009 (Ink on paper, 7.25 x 11")

The visualization of the sound of Charles Bernstein’s recording of “1-100” (1969), which I presented in a recent commentary titled “Anti-ordination in the visualization of the poem's sound,” struck artist, poet, maker of books Stephen Vincent as interestingly relevant to “haptic” drawings he has made while listening to various poets reading their work in the Bay Area, and I agree. He has called this activity drawing by sound (rather than of). “I like comparing my ‘physio/digital’ responses to the digital electronic ones,” he has written to me.

Anti-ordination in the visualization of the poem's sound

Bernstein chants 73 through 75 in "1 to 100" (1969)

ARLO visualization of the PennSound recording of Charles Bernstein's "1 to 100," numbers 73 through 75

Through ARLO (Adaptive Recognition with Layered Optimization), enabled by the HiPSTAS (High Performance Sound Technologies for Access and Scholarship) project headquartered at the Information School of the University of Texas at Austin, I sought to visualize the later passages of Charles Bernstein's chanted/screamed list or counting poem, “1 to 100” (1969). Thanks to Chris Mustazza, Tanya Clement, David Tcheng, Tony Borries, Chris Martin, and others, I am finally learning how to use ARLO to some rudimentary effect. Every single PennSound recording is now available in a test space to which ARLO can be applied by researchers, including myself, associated with the project. We are just beginning. HiPSTAS has received two NEH grants to make all this possible, and PennSound is a founding archival partner.

Syndicate content