The information of consonant factor exists in a higher frequency region, i.e., 7 to 8 kHz[14] 14].
Third, dissatisfied patients have greater need for information which is consonant with other results.
In fact, the parts containing unvoiced consonants do not carry any information about emotions1, thus significantly reducing the complexity of the input signals.
In particular, variations in the gap duration (e.g., the voice onset time) for consonant-vowel combinations convey important information, which determine whether people hear, for example, the phoneme \ta\ or the phoneme \da\.
In addition, the system separates non-sonorant consonants into three classes based on the spectral information (slope and peak location) of fricative consonants.
One analysis of these facts is that vowel-initial words actually begin, at an abstract level of representation, with a kind of "empty" consonant that consists of nothing except the information "broad" or "slender".
For consonant recognition, patterns of relative contributions of spectral and temporal information were similar to those in normal hearing subjects.
Adjacent to broad consonants it is usually a mid central, e.g. [ˈoːl̪ˠəsˠ] "information", but when the preceding syllable contains one of the close back vowels /uː, ʊ/, it is realized as a mid-centralized back, e.g. [ˈd̪ˠuːn̪ˠʊ̽] "closing", [ˈmˠʊkʊ̽] "pigs".
Additional classificatory information may indicate whether the airstream powering the production of the consonant is from the lungs (the pulmonary airstream mechanism) or some other airstream mechanism and whether the flow of air is ingressive or egressive.
Figure 14 shows the confusion matrix for all the consonant-vowels syllables, for the bimodal synthetic condition, where a more detailed information about the confusability among the syllables is presented.
