soundgen (version 2.7.2)

getEntropy: Entropy

Description

Returns Weiner or Shannon entropy of an input vector such as the spectrum of a sound. Non-positive input values are converted to a small positive number (convertNonPositive). If all elements are zero, returns NA.

Usage

getEntropy(
  x,
  type = c("weiner", "shannon")[1],
  normalize = FALSE,
  convertNonPositive = 1e-10
)

Arguments

x

vector of positive floats

type

'shannon' for Shannon (information) entropy, 'weiner' for Weiner entropy

normalize

if TRUE, Shannon entropy is normalized by the length of input vector to range from 0 to 1. It has no affect on Weiner entropy

convertNonPositive

all non-positive values are converted to convertNonPositive

Examples

# Here are four simplified power spectra, each with 9 frequency bins:
s = list(
  c(rep(0, 4), 1, rep(0, 4)),       # a single peak in spectrum
  c(0, 0, 1, 0, 0, .75, 0, 0, .5),  # perfectly periodic, with 3 harmonics
  rep(0, 9),                        # a silent frame
  rep(1, 9)                         # white noise
)

# Weiner entropy is ~0 for periodic, NA for silent, 1 for white noise
sapply(s, function(x) round(getEntropy(x), 2))

# Shannon entropy is ~0 for periodic with a single harmonic, moderate for
# periodic with multiple harmonics, NA for silent, highest for white noise
sapply(s, function(x) round(getEntropy(x, type = 'shannon'), 2))

# Normalized Shannon entropy - same but forced to be 0 to 1
sapply(s, function(x) round(getEntropy(x,
  type = 'shannon', normalize = TRUE), 2))