Ondrej Such

Learn More
In our work we define a new algebra of operators as a substitute for fuzzy logic. Its primary purpose is for construction of binary dis-criminators for phonemes based on spectral content. It is optimized for design of non-parametric computational circuits, and makes uses of 4 operations: min, max, the difference and generalized additively homogenuous means.
Estimating class membership probabilities is an important step in many automated speech recognition systems. Since binary classifiers are usually easier to train, one common approach to this problem is to construct pair-wise binary classifiers. Pairwise models yield an over-determined system of equations for the class membership probabilities. Motivated by(More)
We consider the ability of a very simple feed-forward ne-ural network to discriminate phonemes based on just relative power spectrum. The network consists of two neurons with symmetric nonlinear response over a spectral range. The output of the neurons is subsequently fed to a comparator. We show that often this is enough to achieve complete separation of(More)
  • 1