Skip to main content

Table 2 Audio features used in soundnail creation.

From: Semantic Labeling of Nonspeech Audio Clips

Feature Description
Mean and standard deviations of RMS Energy The average frequency that will vary for each signal.
Spectral Centroid The average frequency of the signal weighted by magnitude.
Spectral Flux How much the frequency varies over time.
50% and 80% Spectral Rolloff How much of the frequencies are concentrated below a given threshold (50% and 80%).
6–10 MFCCs Mel-Frequency Cepstral Coefficients: amplitudes of spectra specified by a set of filters.