APPLICATION OF BLIND SOURCE SEPARATION IN SPEECH PROCESSING FOR COMBINED INTERFERENCE REMOVAL AND ROBUST SPEAKER DETECTION USING A TWO-MICROPHONE SETUP
Erik Visser, Te-Won Lee · 2015
A speech enhancement scheme is presented integrating spatial and temporal signal processing methods for blind denoising in non sta-tionary noise environments. In a first stage, spatially localized in-terferring point sources are separated from noisy speech signals recorded by two microphones using a Blind Source Separation (BSS) algorithm assuming no a priori knowledge about the sources involved. Spatially distributed background noise is removed in a second processing step. Here, the BSS output channel contain-ing the desired speaker is filtered with a time-varying Wiener fil-ter. Noise power estimates for the filter coefficients are computed from desired speaker absent time-intervals identified by compar-ing signal energy of separated source files from the BSS stage. The scheme’s performance is illustrated by speech recognition ex-periments on real recordings corrupted by babble noise and com-pared to conventional beamforming and single channel denoising techniques. 1.