Combining Neural Networks and Hidden Markov Models for automatic detection of pathologies

In the current panorama the conclusive identification of a laryngeal pathology relies inevitably on the observation of the vocal folds by means of laryngoscopical techniques. This inspection technique is inconvenient for a number of reasons,such as its high cost, the duration of the inspection, and. above all, the, fact that it is an invasive technique. This paper looks into a voice recognition system which allows the automatic detection of dysfunction in phonation. The voice signal is parametrized by means of the classic parameters (Hitter, Shimmer, Energy Balance, Spectral Distance) and new Hi;h Order Statistics (HOS) based parameters. The classifier is based in combining of Hidden Markov Models (HMM) and Neural Networks (NN).