EURASIP Journal on Audio, Speech, and Music Processing

Table 4 Comparison between different pooling methods

From: Speech emotion recognition based on Graph-LSTM neural network

Pooling	Max	Sum	Mean	TopK	Weighted
UA (%)	57.97	58.55	59.16	61.88	65.39
WA (%)	66.15	67.71	68.15	68.49	71.83

Back to article page