Energy-based speech enhancement technique for hands-free communication

A dual channel speech enhancement technique for hands-free applications is presented. The proposed method is based on the difference between the powers of two received signals in near-field condition. The subtraction of power spectral densities of noisy signals at two microphones is employed as a criterion for noise reduction. Obtained results show the capability of the proposed method in various noisy environments.

[1]  Régine Le Bouquin-Jeannès,et al.  Enhancement of speech degraded by coherent and incoherent noise using a cross-spectral estimator , 1997, IEEE Trans. Speech Audio Process..

[2]  Stanley T. Birchfield,et al.  Acoustic localization by interaural level difference , 2005, Proceedings. (ICASSP '05). IEEE International Conference on Acoustics, Speech, and Signal Processing, 2005..

[3]  Gérard Faucon,et al.  Using the coherence function for noise reduction , 1992 .

[4]  Parham Aarabi,et al.  Phase-based dual-microphone robust speech enhancement , 2004, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).