Become an expert in R — Interactive courses, Cheat Sheets, certificates and more!
Get Started for Free

logspec.dist

Log-spectral distance

Description

Compare two distributions (e.g. two frequency spectra) by computing the log-spectral distance

Usage

logspec.dist(spec1, spec2, scale=FALSE)

Arguments

spec1

any distribution, especially a spectrum obtained with spec or meanspec (not in dB). This can be either a two-column matrix (col1 = frequency, col2 = amplitude) or a vector (amplitude).
spec2

any distribution, especially a spectrum obtained with spec or meanspec (not in dB). This can be either a two-column matrix (col1 = frequency, col2 = amplitude) or a vector (amplitude).
scale

a logical, if TRUE the distance is scaled by dividing by the square-root of the length of spec1 (or spec2).

Details

The distance is computed according to:

D(spec1 || spec2) = D(spec2 || spec1) = sqrt(sum((10*log10(spec1/spec2))^2))


If scale = TRUE the distance is divided by the length of spec1 (or spec2).

Value

A numeric vector of length 1 returning the D distance.

Note

The function works for both Hz and (htk-)mel scales.

Note

The distance is symmetric.

Author(s)

Jerome Sueur, improved by Laurent Lellouch

See Also

Examples

# Comparison of two spectra
data(tico)
tico1 <- spec(tico, at=0.65, plot=FALSE)
tico2 <- spec(tico, at=1.1, plot=FALSE)
logspec.dist(tico1, tico2)
logspec.dist(tico1, tico2, scale=TRUE)
seewave

Sound Analysis and Synthesis

v2.1.6
GPL (>= 2)
Authors
Jerome Sueur <sueur@mnhn.fr> [cre, au], Thierry Aubin [au], Caroline Simonis [au], Laurent Lellouch [main ctrb], Ethan C. Brown [ctrb], Marion Depraetere [ctrb], Camille Desjonqueres [ctrb], Francois Fabianek [ctrb], Amandine Gasc [ctrb], Eric Kasten [ctrb], Stefanie LaZerte [ctrb], Jonathan Lees [ctrb], Jean Marchal [ctrb], Andre Mikulec [ctrb], Sandrine Pavoine [ctrb], David Pinaud [ctrb], Alicia Stotz [ctrb], Luis J. Villanueva-Rivera [ctrb], Zev Ross [ctrb], Carl G. Witthoft [ctrb], Hristo Zhivomirov [ctrb].
Initial release
2020-06-28

We don't support your browser anymore

Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.