Subject: Re: soundscape analysis? Get real!
From: John Young (John.Young@vuw.ac.nz)
Date: Wed Dec 15 1999 - 19:06:20 EST
Big dangers in attempting to correlate spectrograms/sound with words/text,
not least because reading a language is quite precise compared to
spectrographic representation. Sure, paralinguistic inflexion is always
going to add to or shade meaning of text, but that's a different matter to
trying to interpret the shape of a spectrogram as sound. It's pretty clear
that a spectrogram can reveal interesting things about the way a sound
works for us, but really only in conjunction with hearing itself. Details
of harmonic/inharmonic relationship, which may support or hinder timbral
fusion for instance. But don't forget that the FFT is really an
approximation, skewed by the artifact of its implementation.
Perhaps this distorts the original intention of Carlos's posting, since he
was only after a tool to do a job (probably a purely descriptive one)! But
*analysis* of soundscapes and EA must surely require first and foremost
conceptual tools as, in fact, does analysis of instrumental music - which
is what I think Rob W was getting at.
Primauté de l'oreille.
>Hmmmmm we use graphics to perceive and analyse words. While text is not
>'graphic' in the traditional sense, it appears to me that I use many of
>my senses to figure out what is going on around me. I would not limit
>myself to my ears for understanding sound anymore than I would limit
>myself to only looking at and reading about Joyce. I would consider the
>'rejection' of a mode of communication / understanding, to be limiting on
>my ability to perceive and make sense of an external reality. (If such a
>And this morning 5 cm of snow
Dr John Young Email: John.Young@vuw.ac.nz
Director of Electroacoustic Music Studios Voice: +64 4 463-5864
School of Music Fax: +64 4463-5157
Victoria University of Wellington
PO Box 600, Wellington
This archive was generated by hypermail 2b27 : Wed Jun 11 2003 - 13:09:17 EDT