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 4 Structure of the feature set P4

Number	Name	Type	Frame	Value
1	F12 position ratio	Basic spectral	Voiced	Mean
2	F12 position ratio	Basic spectral	Voiced	Std
3	F12 formant tilt	Basic spectral	Voiced	Min
4	HNR	Complementary spectral	Voiced	Mean
5	HNR	Complementary spectral	Voiced	Std
6	SC	Complementary spectral	Voiced	Mean
7	SC	Complementary spectral	Voiced	Std
8	SFM	Complementary spectral	Voiced	Mean
9	SFM	Complementary spectral	Voiced	Std
10	SE	Complementary spectral	All	Mean
11	SE	Complementary spectral	All	Std
12	F0	Supra-segmental	Voiced	Median
13	F0_DIFF	Supra-segmental	Voiced	Rel. max
14	Jitter	Supra-segmental	Voiced	Median
15	Jitter	Supra-segmental	Voiced	Rel. max
16	Shimmer	Supra-segmental	All	Median