EURASIP Journal on Audio, Speech, and Music Processing

Table 7 Accuracy rates of dialogue act detection with various feature sets in various noisy conditions

From: Robust dialogue act detection based on partial sentence tree, derivation rule, and spectral clustering algorithm

	40%-SIM	60%-SIM	Football	Human	Both	84.8%-ASR	100%-REF
baseline	17.2	32.6	44.3	43.1	42.6	49.6	60.9
NEC	22.4	36.8	52.1	51.0	49.8	56.8	76.9
PS	29.8	49.2	75.2	74.6	73.5	76.2	91.1
uwDR	26.3	48.0	81.1	80.8	80.2	81.6	92.1
DR	26.3	47.4	82.3	81.9	81.6	82.9	93.3

Back to article page