SPECTRUM-BASED ADAPTIVE CANCELLER OF ACOUSTIC ECHOES ARISING IN HANDS-FREE AUDIO

The present echo canceller utilizes the principle that the spectrum pattern of human speech does not change much in the short run. The inputs to the present echo canceller are x(t) and y(t), y(t) representing the incoming speech signal from a far-end speaker and x(t) representing the combination of speech signal from a near-end speaker and the echo. The processed forms of the input x(t) and y(t) are processed by applying the well-known Hanning window. They are then transformed into their respective frequency domain using the well-known fast Fourier transform (FFT), and the power spectrum P.sub.x and P.sub.y are calculated where EQU P.sub.x =.vertline.x.sub.r (f).vertline.+.vertline.x.sub.i (f).vertline.+.epsilon.*.vertline.x.sub.r (f).vertline.*.vertline.x.sub.i (f).vertline. and EQU P.sub.y =.vertline.y.sub.r (f).vertline.+.vertline.y.sub.i (f).vertline.+.epsilon.*.vertline.y.sub.r (f).vertline.*.vertline.y.sub.i (f).vertline. where .epsilon. is a scaling factor which controls the amount of echo to be suppressed, and converting P.sub.x and P.sub.y to bark scales P.sub.x (b) and P.sub.y (b). The transfer function H(b) is then estimated using the Bark Scales. The transfer function is used to normalize P.sub.y (b), which, in turn together with P.sub.x (b), is: used to estimate the gain G(b) which will be used to suppress the echo. Subsequently, the Bark Scales are unwarped and the gain function is then used to suppress the echo from the input x(t). The well-known inverse FFT (IFFT) and overlap add are performed to yield an echo-free signal.