Phoneme discrimination using $KS$-algebra II
This work addresses phoneme discrimination for speech processing, but it appears incremental as it applies an existing algebraic framework to a specific domain.
The paper tackled binary classification of English phonemes using $KS$-algebra-based $Z$-classifiers, showing they effectively reflect known vowel formant characteristics while achieving very low Kolmogoroff's complexity.
$KS$-algebra consists of expressions constructed with four kinds operations, the minimum, maximum, difference and additively homogeneous generalized means. Five families of $Z$-classifiers are investigated on binary classification tasks between English phonemes. It is shown that the classifiers are able to reflect well known formant characteristics of vowels, while having very small Kolmogoroff's complexity.