Sixth International Conference on Spoken Language Processing (ICSLP 2000) Beijing, China October 16-20, 2000

Speech/Noise Separation Using Two Microphones and a VQ Model of Speech Signals

Alex Acero, Steven Altschuler (1), Lani Wu (1)

Speech Technology Group, Microsoft Research, Redmond, WA, USA
(1) Presently at Rosetta Impharmatics, Kirkland, WA, USA

In this paper we address the problem of using two or more microphones to enhance speech corrupted by nonstationary noise, such as that of a competing speaker (cocktail party effect) at very low SNR, by means of linear filtering of two microphone signals. This work is a variant to the probabilistic Independent Component Analysis (ICA) method but using a more accurate probability distribution of the speech signal based on a mixture Autoregressive model. Comparison with other algorithms on published real recordings shows more separation than with previously existing ICA algorithms.

Full Paper

Bibliographic reference.  Acero, Alex / Altschuler, Steven / Wu, Lani (2000): "Speech/noise separation using two microphones and a VQ model of speech signals", In ICSLP-2000, vol.4, 532-535.