Skip to main content
. 2021 Mar 22;8(21):16035–16046. doi: 10.1109/JIOT.2021.3067605

Fig. 3.

Fig. 3.

Scheme of the statistical functionals approach. The frame-level LLDs (e.g., MFCCs) are first extracted from the speech signal. Then, a series of statistical functionals (e.g., max., min., mean, etc.) can be calculated from these LLDs, as scalars independent of the length of the instances.