Biomedical Engineering Reference
In-Depth Information
(a)
10000
0.0
0.0
time (s)
1.1
(b)
5000
0.0
0.0
time (s)
1.0
Fig. 5.1.
Sonograms of two song fragments: (
a
) one vocalized by an ashy-tailed
swift (
Chaetura andrei
), and (
b
) one by a greenish manakin (
Schiffornis virescens
).
The songs are globally very different. The timbres of the sounds might differ, as
well as the average fundamental frequency of the notes. The combinations of sounds
used to compose the song are not the same. However, at the level of the note, there
is one important similarity: in both cases the sonogram shows a sequence of small
continuous “curves”. Each continuous curve sweeps a certain frequency range. The
frequency typically evolves within a syllable, either upwards, downwards or “
n
”
shaped
Calder conjectured that the spaces between syllables are used to make
the mini-inspirations needed to execute the song [Calder 1970]. This hy-
pothesis was experimentally validated by Hartley [Hartley and Suthers 1989,
Hartley 1990]. She found that each syllable is accompanied by an air pulse
(expiration), while during each intersyllabic silence, the air sac pressure falls
below atmospheric pressure (inspiration) creating an inflow of air. This res-
piratory strategy is not used in trills repeated at high frequencies, but in
canaries, the existence of mini-inspirations was found in executions of more
than 30 syllables per second.
The execution of a syllable after a mini-inspiration implies a raising of
the air sac pressure. This increase in pressure in the air sac continues until it
reaches a value such that, according to our discussion in the previous chapter,
the force exerted by the interlabial pressure overcomes the dissipative forces.
In this way, the labial oscillations that generate the sound are established. Af-
ter having sustained these oscillations for a while, the pressure decreases and
