Biomedical Engineering Reference
In-Depth Information
10000
0.0
15.0
time (s)
20.0
Fig. 1.8.
A complex sonogram. This is the sonogram of a grass wren's song. Here
you can see
upsweeps
(increasing fundamental frequency),
downsweeps
(decreasing
fundamental frequency) and
tonal
sounds (constant frequency). The rich vocaliza-
tions of this wren span a huge range in frequency, from around 2000 to 8000 Hz
In Fig. 1.7b, we see the results of the procedure. The time is on the hori-
zontal axis, while the vertical axis indicates the value of the frequency. The
fact that the curve is an almost horizontal stroke indicates that the sound,
within this brief syllable, has an almost constant frequency - a note. This type
of representation is known as a
sonogram
. Of course, instead of focusing only
on the fundamental frequency, one can take into account the fundamental
and all other frequencies that appear in the spectrum in each time window.
The resulting sonogram allows us not only to track the time evolution of
the fundamental frequency but also to have a picture of the complete spec-
trum, evolving in time. In very complex songs, other frequencies appearing in
the sonogram may have an evolution different from that of the fundamental
frequency.
In Fig. 1.8, we show a sonogram corresponding to the song of the grass
wren (
Cistothorus platensis
) [Straneck 1990a]. Notice that the structure of
this song is extremely complex: syllables are repeated in rapid succession
before being replaced by others that are qualitatively different. Some are
upsweeps (the fundamental frequency increases), some are downsweeps, and
others are tonal sounds (and therefore could have been described as simple
notes). In this topic, we intend to provide an understanding to understand
what physical processes are at play in generating such a variety of structures.
In the process of introducing some elements for the description of these phe-
nomena, we have discussed this graphical representation here, which will
allow us to represent the songs that we hear, diminishing the ambiguity of
the onomatopoeic description.
