Supervised speech separation combined with adaptive beamforming

Authors:           Zoran Šarić1, Miško Subotić1, Ružica Bilibajkić1, Marko Barjaktarović2, Nebojša Zdravković3

 

Laboratory of Acoustics, Life Activities Advancement Center,

Gospodar Jovanova 35, 11000 Belgrade, Serbia1

 

Faculty of electrical engineering University of Belgrade, Bulevar kralja Aleksandra 73, 11120 Belgrade, Serbia2

 

Faculty of medical sciences University of Kragujevac, Svetozara Markovica 69,

34000 Kragujevac, Serbia3

 

 

 

 

·         How to install supervised speech separation-Adaptive beamforming

·         Download DNN-ABF matlab code

 

 

Test scenario, T60 = 360 ms

 

Test examples

 

ratio_1_clean.wav                           Clean speech of the uterence 1

ratio_1_refMic.wav                        Noisy speech of the uterence 1 recorded on microphone 4

ratio_1_estimated.wav                 Speech Processed by supervised speech separation (SSS)

ratio_1_gsc.wav                               Speech processed by S-GSC

ratio_1_gscPost2.wav                    Speech processed by postfilter