Dear Jiajun,
There have been numerous clues that attack transients of musical
instrument tones are important, but there's not been a lot of hard
evidence.
The 1963 paper by Clark, et al.:
"Preliminary Experiments on the Aural Significance of Parts of Tones
of Orchesteral Instruments and on Choral Tones", J Audio Eng Soc,
Vol 11, no. 1, pp 45-54 (1963)
gives confusions matrices for 13 sustain tone instruments for normal
tones, first 0.15 s of tones, first 0.06 s of tones, and 0.60 s of
their "steady states".
These matrices showed that "identification of nonpercussive orchestral
instruments from the first 0.15 sec of their tones is not much worse
than from the normal tone". It was "very striking that the
identifications
(were) so very resistent to shortening the duration of the note".
It would be good to see more evidence about the importance of
transients
for identification. I suspect that much shorter transients could be
identified for percussive instruments. E.g., piano vs. guitar vs. koto.
Jim Beauchamp
University of Illinois
Original message:
From: Jiajun Yang <thejyang@xxxxxxxxx>
Date: Fri, 16 Mar 2018 11:24:33 +0100
To: AUDITORY@xxxxxxxxxxxxxxx
Subject: Shortest duration to perceive timbre or tonal quality?
Suggest reference
Dear colleagues,
I am hoping if you can point us to some studies on whether there is a
minimum duration of sound for its timbre or tonal quality to be
clearly
perceived.
I found this quote from
http://www.indiana.edu/~emusic/etext/acoustics/chapter1_timbre.shtml
<http://www.indiana.edu/~emusic/etext/acoustics/chapter1_timbre.shtml>
that "Some studies have indicated it takes at least 60ms to recognise
the timbre of a sound" However, I am not about to locate the exact
study.
Many thanks
Jiajun Yang
Postdoc research, Ambient Intelligence Group,
Citec, Bielefeld University, Germany=