[From Bruce Nevin (Thu 930520 14:30:21)]
Some interesting experimental data about speech:
The difference between syllables /pa ta ka/, or between syllables
/ti te ta to tu/, is in the transition between silence (when the
tongue blocks the air stream) and the vowel.
Background, skip if familiar:
The acoustic signal for a vowel has energy at the fundamental
frequency (pitch) of the voice, that is, the rate of vibration of the
vocal folds, and in ranges of higher frequencies, for those harmonics
of the fundamental that are not damped by the air passage between the
larynx and the lips. Different frequencies are damped, depending upon
the configuration of the tongue and lips (and to a certain extent other
articulators). As the articulators move from closing the air passage
to (or toward) the configuration for the vowel, the formants may rise or
fall in pitch. The same consonant may appear in the acoustic signal
as an up glissando with one vowel, a down glissando with another.
Different consonants before a given vowel are differentiated in the
acoustic signal by different formant glissandi in the transition between
consonant and vowel. There are complications, but this is a good enough
first approximation.
If you clip the transition between silence and vowel from one syllable,
and substitute it for that for another, you hear the first syllable.
If you listen to the transitions alone, you hear only chirps and
swift glissandi. Sounds like maybe snippets of birdsong. Certainly
sounds nothing like speech.
Experiment 1: In one ear (say, the right ear) play the acoustic signal
for syllables, only with gaps (silence) in place of the transitions.
In the other ear (the left) play the glissandi for various syllables,
appropriately timed to match the gaps.
Experiment 2: play the syllable /ba/ three times while displaying
a face pronouncing /bE vE thE/ perfectly synchronized with the
acoustic signal. (/E/ is the vowel in "bed", /dh/ is the consonant
in "the".).
We pause here for you to guess what you would perceive with these
experimental setups. In experiment 1, what do you hear in the right ear
and what do you hear in the left? Are there changes as amplitude is
decreased equally in both channels, and if so, what? In experiment 2,
what do you hear in the headphones? These experiments have been done,
so I can report the results.
Bruce
bn@bbn.com