EURASIP Journal on Audio, Speech, and Music Processing

Table 6 The estimation performances for TURES and VAM databases

From: Recognizing emotion from Turkish speech using acoustic features

		Valence			Activation			Dominance
Train	Test	CC	MAE	RMSE	CC	MAE	RMSE	CC	MAE	RMSE
TURES	TURES	0.288	0.166	0.232	0.739	0.156	0.201	0.743	0.119	0.153
VAM	TURES	0.131	0.204	0.275	0.624	0.191	0.246	0.649	0.165	0.208
VAM	VAM	0.310	0.149	0.196	0.810	0.160	0.202	0.761	0.153	0.196
TURES	VAM	0.289	0.170	0.219	0.743	0.189	0.238	0.717	0.178	0.225

Results are given in terms of correlation coefficient (CC), mean absolute error (MAE), and root-mean-square error (RMSE). The cross-corpus results for each primitive are also given.

Back to article page