neural network system comprising a plurality of layers (20, 21, 22).design elements (10, 11) connected thelevels with an entry level (20) in which a sequence of vectorsvoice entry (2) may be applied to at least onethe first pace. at least one element (10) includes a decimator(102) in order to reduce the frequency of the signal received. athese vectors can deliver paramu00e9triseur voice (2) from athe digital voice signal. voice recognition for benefits:based on a set of vectors corresponding to a number of voice framesthe voice signal.
展开▼
