and 
for vowel sounds.
The sounds of speech are generated by the speech mechanism of a vast
variety of different speakers, even if we confine our attention to the
sounds of native speakers in a specific language. All these speakers
have a larynx and vocal tract of different dimensions, they have
different speech habits and a voice quality. Even one individual
speaker talking at different times will produce sounds which are
acoustically very different so that any utterance we may take for
purposes of acoustic analysis can be only one sample out of millions
of possible ones[30]. However, despite acoustic variability,
it is both useful and meaningful to make generalizations about the
acoustic characteristics of the different classes of sound that make
up a target linguistic sound system. As criteria to feedback
instruction, we need to set up articulatory statistical models by
native speakers to represent the variability in formants and the
spectral features in terms of both the place of articulation and the
manner of articulation. For the word utterances collected from 48
native-male speakers, we have trained Pair-Wise classifiers from their
consonant sounds as well as set up the distributions of their
and for vowel sounds.
, ) for Vowels
