Characterizing audio signals

摘要: The loudness, spectral mean and spread of speech signals are represented in the visual domain similar to brightness, hue saturation a color, respectively. above parameters signal extracted, and, by various operations, adapted for use in, and/or with other systems. As values these defined relative reference frames such that so extracted large degree insensitive extraneous ambient noises, speaker differences overall (wideband) filterings.