Yaron Laufer and Sharon Gannot, submitted to EUSIPCO 2020.

T₆₀ =160ms, SNR=0dB:

FLMC0+MCEW0

Noisy Mixture	Clean - Speaker 1	Clean -Speaker 2
	ILRMA [1]- speaker 1	ILRMA [1]- speaker 2
	Proposed- speaker 1	Proposed - speaker 2

T₆₀ =160ms, SNR=5dB:

FCYL0+MMDS0

Noisy Mixture	Clean - Speaker 1	Clean -Speaker 2
	ILRMA [1]- speaker 1	ILRMA [1]- speaker 2
	Proposed- speaker 1	Proposed - speaker 2

T₆₀ =360ms, SNR=0dB:

FAJW0+MPPC0

Noisy Mixture	Clean - Speaker 1	Clean -Speaker 2
	ILRMA [1]- speaker 1	ILRMA [1]- speaker 2
	Proposed- speaker 1	Proposed - speaker 2

T₆₀ =360ms, SNR=5dB:

FTMG0+MCTM0

Noisy Mixture	Clean - Speaker 1	Clean -Speaker 2
	ILRMA [1]- speaker 1	ILRMA [1]- speaker 2
	Proposed- speaker 1	Proposed - speaker 2

References:

[1]  D. Kitamura, N. Ono, H. Sawada, H. Kameoka, and H. Saruwatari, “Determined blind source separation unifying independent vector analysis and nonnegative matrix factorization,” IEEE/ACM Trans. Audio, Speech, Lang. Process., vol. 24, no. 9, pp. 1626–1641, 2016.