Linguistics Encyclopedia.pdf

The linguistics encyclopedia 6
articulate the vowels [i], [ ] and [u] without varying the size of the oral cavity (see also
ARTICULATORY PHONETICS). What does this mean acoustically?
For the sake of illustration, let us assume that a tube consisting of the joined oral and
pharyngeal cavities is a resonating acoustic tube, much like an organ pipe. The most
uniform ‘pipe’ or tube one can assume is the one formed when producing the neutral
vowel [ ] (see Figure 7). Without going into much detail, the natural resonance
frequency of such a tube can be calculated with the following formula:
Where f=frequency, v=velocity of sound, and l=length of the vocal tract
Since v is 340 m per second, and l is 17 cm in an average male, f is about 500 Hz when
n=1,1,500 Hz when n=2,2,500 Hz when n=3, etc. What
Figure 7 The vocal-tract shape and an
idealized tube model of the tract for the
most neutral vowel
this means is that, given a vocal tract which is about 17 cm long, forming the most
neutral tract shape usually assumed for the schwa vowel [ ], the white noise (the vocalfold
excitation) at one end will be modified in such a way that there will be resonance
peaks at every 1,000 Hz beginning at 500 Hz. These resonance peaks are called
formants.
It is easy to imagine that a change in the size and shape of a resonating acoustic tube
results in the change of resonance frequencies of the tube. For the purpose of speech
acoustics, it is convenient to regard the vocal tract as consisting of two connected tubes,
one front and the other back with the velic area as the joint. Viewed in this way, vowel [i]
has the narrow front (oral) tube and the wide back tube, while [ ] is its mirror image,
i.e, [ ] has the wide front tube but the narrow back tube. On the other hand, [u] has the
narrow area (‘the bottle neck’) in the middle (at the joint) and, with the lip rounding, at