Comparing Gaussian and polynomial classification in SCHMM-based recognition systems

Semi-continuous Hidden Markov Models (SCHMM) with gaussian distributions are often used in continuous speech or handwriting recognition systems. Our paper compares gaussian and tree-structured polynomial classi ers which have been successfully used in pattern recognition since many years. In our system the binary classi er tree is generated by clustering HMM states using an entropy measure. For handwriting recognition, gaussians are clearly outperformed by polynomial classi cation. However, for speech recognition, polynomial classi cation currently performs slightly worse because some system parameters are not yet optimized.