[Retracted] Improvement of Speech Recognition Technology in Piano Music Scene Based on Deep Learning of Internet of Things

Table 3

Sparsity required for different DeSNRs.

DeSNR	0.4	1.3	1.8	2.1	2.6	3.1	4.1	4.7	5.2	6.2	7.4
Sparsity	10	100	200	300	500	900	2800	5800	8800	19K	40 K
Decomposition time/s	0.15	0.17	0.24	0.29	0.79	1.68	4.82	9.51	13.2	30.4	61.7