Skip to main content


Table 3 Sixty-four low-level descriptors (LLD) of the ComParE feature set

From: Emotion in the singing voice—a deeperlook at acoustic features in the light ofautomatic classification

Four energy-related LLD
Sum of auditory spectrum (loudness)
Sum of RASTA-style filtered auditory spectrum (modulation loudness)
RMS energy, zero-crossing rate
Fifty-five spectral LLD
RASTA-style auditory spectrum, bands 1–26 (0–8 kHz)
MFCC 1–14
Spectral energy 250–650 Hz, 1 k–4 kHz
Spectral roll off points 0.25, 0.50, 0.75, 0.90
Spectral flux, centroid, entropy, slope
Variance, skewness, kurtosis
Psychoacoustic sharpness and harmonicity
Six voicing-related LLD
F 0 via sub harmonic summation (SHS) and Viterbi smoothing
Probability of voicing, logarithmic HNR by waveform matching
Jitter (local and delta), shimmer (local)