Process for voice recognition in a noisy acoustic signal and system implementing this process

Abstract

The invention relates to a process and a system for voice recognition in a noisy signal. In a preferred embodiment, the system () comprises modules for detecting speech () and for formulating a noise model (), a module () for quantifying the energy level of the noise and for comparing with preestablished energy spans, a parameterization pathway () comprising an optional denoising module (), with Wiener filter, a module () for calculating the spectral energy in Bark windows, a module () for applying a configuration of shift values (), by adding these values to the Bark coefficients, as a function of the quantification (), so as to modify the parameterization, a module () for calculating vectors of parameters, and a block () for recognizing shapes, performing the voice recognition by comparison with vectors of parameters prerecorded during a learning phase.