Evaluation of influence of spectral and prosodic features on GMM classification of Czech and Slovak emotional speech

EURASIP Journal on Audio, Speech, and Music Processing

Table 2 Structure of the feature set P2

Number	Name	Type	Frame	Value
1	Spectral envelope	Basic spectral	Voiced	Skewness
2	Spectral envelope	Basic spectral	Voiced	Kurtosis
3	SC	Complementary spectral	Voiced	Min
4	Spectral spread	Basic spectral	Voiced	Std
5	SFM	Complementary spectral	Voiced	Mean
6	Spectral decrease	Basic spectral	Voiced	Min
7	F0	Supra-segmental	Voiced	Std
8	F0	Supra-segmental	Voiced	Rel. max
9	F0_DIFF	Supra-segmental	Voiced	Median
10	F0_DIFF	Supra-segmental	Voiced	Std
11	F0_ZCR	Supra-segmental	Voiced	Median
12	F0_ZCR	Supra-segmental	Voiced	Rel. max
13	Jitter	Supra-segmental	Voiced	Median
14	Jitter	Supra-segmental	Voiced	Rel. max
15	Shimmer	Supra-segmental	All	Median
16	Shimmer	Supra-segmental	All	Rel. max