EURASIP Journal on Audio, Speech, and Music Processing

Table 3 Performance comparison of different models with and without the application of SpecAugment

From: Exploring the power of pure attention mechanisms in blind room parameter estimation

Method	Dataset I				Dataset II
Method	MSE	MAE	\(\varvec{\rho}\)	MM	MSE	MAE	\(\varvec{\rho}\)	MM
CNN [21]	0.3863	0.4837	0.6984	3.0532	0.3154	0.4136	0.7678	2.5921
CRNN	0.3572	0.4265	0.7262	2.6701	0.2818	0.3684	0.7898	2.3471
Proposed method	0.2650	0.3432	0.8077	2.2039	0.1981	0.2884	0.8580	1.9427
Proposed method w/ pretrain	0.2157	0.3111	0.8529	2.0470	0.1541	0.2423	0.8929	1.7470

Back to article page