A Speech Understanding System With Learning Capability

Abstract

A speech understanding aystem w ith learning capabilities Is presented, lis relevant aspects are : a) the spoken sentence Is repre&ented conc i sely by a description lhal can be used 10 reconstruct the sentence and to verify whether its meaning was not degraded by the coding. b} Syll•blea or bro•der coarliculatlon segnents ef"e the smalleat units . c) The evaluation of an hypothes is Ia baaed 011 the probabil ity that a syllable. a word or the sentence can gener ate the spectrogram of the spoken message and th at a syntact ic structure can gener11te Its pitch contour . d) Coartic:ulation effects are described in terms of patlern grammars, ~~nera t :n g all the possib le formant trajectories for a g ive n ullerance . e) Spectral and prosodic featu res c an be learned by In ference of stoch~stlc f inlte- state-automata. f) More for,ant choices are allowed for a single syllable segment .. and an algof'lttwn Ia provided for ass i!J"ing to each cholco e probability.