PHONETIC KNOWLEDGE EMBEDDED IN A CONTEXT-SENSITIVE MLP FOR FRENCH SPEAKER-INDEPENDENT SPEECH RECOGNITION

The stop /p,t,k/ recognition part of a speaker-independent speech reco gnition system is described in this paper. This work is based on the c onclusions of several perceptual experiments and on the results of an acoustic investigation with stop consonants. These experiments allowed us to evaluate the discrimination power of the burst regarding the st op place of articulation, and how the vocalic information may help sto p identification, which could not be done efficiently without taking i nto account the nature of the following vowel. Thus, a novel system ar chitecture is proposed which is made up of two stages: first, an autom atic detector of reliable cues regarding stop and vowel features, and, then, a context sensitive multilayered perceptron (ODWE) fed by the p revious acoustic cues. The training and the test have been done on two different corpora including male and female speakers. The results sho w a recognition rate of 90% over the stop consonants.