TY - GEN
T1 - Speaker extraction using LCMV beamformer with DNN-based SPP and RTF identification scheme
AU - Malek, Ariel
AU - Chazan, Shlomo E.
AU - Malka, Ilan
AU - Tourbabin, Vladimir
AU - Goldberger, Jacob
AU - Tzirkel-Hancock, Eli
AU - Gannot, Sharon
N1 - Publisher Copyright:
© EURASIP 2017.
PY - 2017/10/23
Y1 - 2017/10/23
N2 - The linearly constrained minimum variance (LCMV)-beamformer (BF) is a viable solution for desired source extraction from a mixture of speakers in a noisy environment. The performance in terms of speech distortion, interference cancellation and noise reduction depends on the estimation of a set of parameters. This paper presents a new mechanism to update the parameters of the LCMV-BF. A new speech presence probability (SPP)-based voice activity detector (VAD) controls the noise covariance matrix update, and a speaker position identifier (SPI) procedure controls the relative transfer functions (RTFs) update. A postfilter is then applied to the BF output to further attenuate the residual noise signal. A series of experiments using real-life recordings confirm the speech enhancement capabilities of the proposed algorithm.
AB - The linearly constrained minimum variance (LCMV)-beamformer (BF) is a viable solution for desired source extraction from a mixture of speakers in a noisy environment. The performance in terms of speech distortion, interference cancellation and noise reduction depends on the estimation of a set of parameters. This paper presents a new mechanism to update the parameters of the LCMV-BF. A new speech presence probability (SPP)-based voice activity detector (VAD) controls the noise covariance matrix update, and a speaker position identifier (SPI) procedure controls the relative transfer functions (RTFs) update. A postfilter is then applied to the BF output to further attenuate the residual noise signal. A series of experiments using real-life recordings confirm the speech enhancement capabilities of the proposed algorithm.
UR - https://www.scopus.com/pages/publications/85041458768
U2 - 10.23919/EUSIPCO.2017.8081615
DO - 10.23919/EUSIPCO.2017.8081615
M3 - Conference contribution
AN - SCOPUS:85041458768
T3 - 25th European Signal Processing Conference, EUSIPCO 2017
SP - 2274
EP - 2278
BT - 25th European Signal Processing Conference, EUSIPCO 2017
PB - Institute of Electrical and Electronics Engineers
T2 - 25th European Signal Processing Conference, EUSIPCO 2017
Y2 - 28 August 2017 through 2 September 2017
ER -