ENHANCEMENT OF A SPEECH SIGNAL EMBEDDED IN NOISY ENVIRONMENT USING TWO MICROPHONES

Allan Kardec Barros 1;5 , Hideki Kawahara 2 , Andrzej Cichocki 3
E­mail: akbarros@ieee.org

In this work we develop a system for enhancement of a speech signal with the highest energy from a linear convolutive mixture of independent sound sources (in­ terferences) recorded by two microphone signals. In this system we use the concept of independent compo­ nent analysis (ICA) together with auditory filter banks, pitch tracking, adaptive band pass filters and masking. Preliminary computer simulations experiments confirm the validity of the proposed algorithm.