Clean speech reconstruction from noisy mel-frequency cepstral coefficients using a sinusoidal model.

Xu Shao, Ben Milner

Browse the full ICASSP paper archive.