Ninth International Conference on Spoken Language Processing

Pittsburgh, PA, USA
September 17-21, 2006

Continuous Time-Frequency Masking Method for Blind Speech Separation with Adaptive Choice of Threshold Parameter Using ICA

Zbynek Koldovsky, Jan Nouza, Jan Kolorenc

Technical University of Liberec, Czech Republic

We propose a novel method for blind speech separation using continuous time-frequency masking. The method is equipped with an adaptive choice of a threshold parameter that is based on utilization of ICA methods. We present a direct application that consists in the speech segregation for automatic transcription of spoken broadcasts disturbed by background music. Experimental results show improved performance in comparison with traditionally used binary masking methods.

Full Paper

Bibliographic reference.  Koldovsky, Zbynek / Nouza, Jan / Kolorenc, Jan (2006): "Continuous time-frequency masking method for blind speech separation with adaptive choice of threshold parameter using ICA", In INTERSPEECH-2006, paper 1224-Thu2FoP.1.