. Author manuscript; available in PMC: 2018 May 1.

Published in final edited form as: IEEE/ACM Trans Audio Speech Lang Process. 2017 Mar 24;25(5):1075–1084. doi: 10.1109/TASLP.2017.2687104

TABLE I.

Average STOI scores (%) of different methods in simulated matched-room and unmatched-room conditions.

	T60	MIX_L	MIX_R	DAS	MVDR	MWF	MESSL	SBC	PRO.
Matched room	0.0s	58.00	58.04	63.56	63.75	66.86	65.92	63.65	74.66
	0.3s	53.13	52.64	58.61	58.78	64.06	58.66	62.79	74.88
	0.6s	44.08	41.00	50.82	50.84	57.72	51.89	55.08	68.53
	0.9s	44.58	43.31	48.20	48.15	57.38	48.46	53.37	65.39

	Avg.	49.05	49.65	55.30	55.38	61.51	56.23	58.72	70.87

Unmatched room	0.2s	55.28	57.20	61.80	61.91	65.35	60.52	64.68	74.95
	0.4s	47.98	48.82	54.46	54.64	61.21	55.91	59.02	70.40
	0.8s	39.99	41.59	47.06	47.01	56.86	47.12	54.27	65.92
	1.0s	39.05	40.95	45.21	45.01	55.55	46.05	51.64	62.82

	Avg.	45.58	47.14	52.13	52.14	59.74	52.40	57.40	68.52