Terminology note: By analogy with Barthes's lexias, I'd like to call those segments of auditory experience into which our hearing of a poem divides "audias" (singular: audia). I supply the following transcript, generated using the "label track" function of Audacity, as a preliminary way of thinking about how sentence, line, and audia interact and inform the analysis of poetry soundfiles. [Back to Lipstick of Noise.]

Lydia Davis - A Position at the University

(Note: time values at left correspond to track w/o commentary; ^ indicates increased emphasis)

00.910822 I think I know what
02.152415 sort of person I am
03.950089 but then I think
04.793798 but this stranger will imagine me quite otherwise when he or she
08.825382 hears this or that to my credit
10.929859 for instance that I have a position at the university

16.040047 the fact that I have a position at the university will appear to mean
19.721684 that I must ^be the sort of person
22.061057 who has a position at the university

25.066768 but then I have to admit with surprise that
27.549956 after all it ^is true that I have a position at the university

32.190352 and if it ^is true then perhaps I really ^am the sort of person you imagine when you
36.509564 hear that a person has a position at the universit

41.250630 but on the other hand I ^know I am ^not the sort of person ^I imagine
44.961029 when I hear that a person has a position at the university

49.265860 then I see what the problem is

51.605233 when others describe me this way they appear to describe me ^completely
55.708724 whereas in fact
56.753772 they do not describe me completely
59.121908 and a complete description of me would include truths
62.386484 that seem quite incompatible with the fact
65.018279 that I have a position at the university