In the recognition phase, the signal originating from a sensor is processed to obtain parameters which are compared with those stored in a dictionary in the learning phases so as to recognize the voice structures uttered by the user in a noisy environment. The obtaining of the said parameters during the learning and recognition phases includes the formation of digital frames of predetermined length from the signal originating from the sensor, the transformation of each frame from the time domain to the frequency domain to obtain a spectrum X(i), and the application of an inverse transformation, from the frequency domain to the time domain, to the magnitude .vertline.X(i).vertline..sup..gamma., where .vertline.X(i).vertline. represents the modulus of the spectrum and .gamma. represents an appropriate exponent.