TABLE III.

Objective scores based on acoustic signals for reverberant (R) algorithm performance. Reverberant target speech was used as the reference.

Input TIR	Unprocessed STOI (%)	Processed STOI (%)	Benefit STOI	Unprocessed PESQ	Processed PESQ	Benefit PESQ
−6 dB	41.51	78.43	36.92	1.59	2.57	0.98
−3 dB	50.42	83.00	32.58	1.73	2.77	1.04
0 dB	59.74	86.83	27.09	1.94	2.97	1.03
3 dB	68.70	89.93	21.23	2.16	3.16	1.00
6 dB	76.66	92.28	15.62	2.41	3.36	0.95
Mean	59.41	86.09	26.69	1.97	2.97	1.00