freq_DTW calculates acoustic dissimilarity of frequency contours using dynamic
time warping. Internally it applies the dtwDist function from the dtw package.
freq_DTW(
X = NULL,
type = "dominant",
wl = 512,
wl.freq = 512,
length.out = 20,
wn = "hanning",
ovlp = 70,
bp = NULL,
threshold = 15,
threshold.time = NULL,
threshold.freq = NULL,
img = TRUE,
parallel = 1,
path = NULL,
ts.df = NULL,
img.suffix = "dfDTW",
pb = TRUE,
clip.edges = TRUE,
window.type = "none",
open.end = FALSE,
scale = FALSE,
fsmooth = 0.1,
adjust.wl = TRUE,
max.obs.per.core = 20,
...
)A matrix with the pairwise dissimilarity values. If img is
FALSE it also produces image files with the spectrograms of the signals listed in the
input data frame showing the location of the dominant frequencies.
object of class 'selection_table', 'extended_selection_table' or data frame containing columns for sound file name (sound.files), selection number (selec), and start and end time of signal (start and end).
Character string to determine the type of contour to be detected. Three options are available, "dominant" (default), "fundamental" and "entropy".
A numeric vector of length 1 specifying the window length of the spectrogram, default is 512.
A numeric vector of length 1 specifying the window length of the spectrogram for measurements on the frequency spectrum. Default is 512. Higher values would provide more accurate measurements.
A numeric vector of length 1 giving the number of measurements of frequency desired (the length of the time series).
Character vector of length 1 specifying window name. Default is
"hanning". See function ftwindow for more options.
Numeric vector of length 1 specifying % of overlap between two
consecutive windows, as in spectro. Default is 70.
A numeric vector of length 2 for the lower and upper limits of a
frequency bandpass filter (in kHz). Default is NULL.
amplitude threshold (%) for frequency detection. Default is 15.
amplitude threshold (%) for the time domain. Use for frequency detection. If NULL (default) then the 'threshold' value is used.
amplitude threshold (%) for the frequency domain. Use for frequency range detection from the spectrum (see 'frange.detec'). If NULL (default) then the
'threshold' value is used.
Logical argument. If FALSE, image files are not produced. Default is TRUE.
Numeric. Controls whether parallel computing is applied. It specifies the number of cores to be used. Default is 1 (i.e. no parallel computing). In this function parallelization improves performance only if the number of rows in 'X' is at least twice the number of cores to be used.
Character string containing the directory path where the sound files are located.
If NULL (default) then the current working directory is used.
Optional. Data frame with frequency contour time series of signals to be compared. If provided "X" is ignored.
A character vector of length 1 with a suffix (label) to add at the end of the names of
image files. Default is NULL.
Logical argument to control progress bar. Default is TRUE.
Logical argument to control whether edges (start or end of signal) in
which amplitude values above the threshold were not detected will be removed. If
TRUE (default) this edges will be excluded and contours will be calculated on the
remaining values. Note that DTW cannot be applied if missing values (e.i. when amplitude is not detected).
dtw windowing control parameter. Character: "none", "itakura", or a function (see dtw).
dtw control parameter. Performs
open-ended alignments (see dtw).
Logical. If TRUE frequency values are z-transformed using the scale function, which "ignores" differences in absolute frequencies between the signals in order to focus the
comparison in the frequency contour, regardless of the pitch of signals. Default is TRUE.
A numeric vector of length 1 to smooth the frequency spectrum with a mean
sliding window (in kHz) used for frequency range detection (when frange.detec = TRUE). This help to average amplitude "hills" to minimize the effect of
amplitude modulation. Default is 0.1.
Logical. If TRUE 'wl' (window length) is reset to be lower than the
number of samples in a selection if the number of samples is less than 'wl'. Default is TRUE.
Numeric. Maximum number of observations per core to be used in parallel computing. Default is 100. Reduce this value if you have memory issues.
Additional arguments to be passed to track_freq_contour for customizing
graphical output.
Marcelo Araya-Salas (marcelo.araya@ucr.ac.cr)
This function extracts the dominant frequency values as a time series and
then calculates the pairwise acoustic dissimilarity using dynamic time warping.
The function uses the approx function to interpolate values between dominant
frequency measures. If 'img' is TRUE the function also produces image files
with the spectrograms of the signals listed in the input data frame showing the
location of the dominant frequencies.
Araya-Salas, M., & Smith-Vidaurre, G. (2017). warbleR: An R package to streamline analysis of animal acoustic signals. Methods in Ecology and Evolution, 8(2), 184-191.
spectrograms for creating spectrograms from selections,
snr_spectrograms for creating spectrograms to
optimize noise margins used in sig2noise and freq_ts, freq_ts, for frequency contour overlaid spectrograms.
Other spectrogram creators:
color_spectro(),
multi_DTW(),
phylo_spectro(),
snr_spectrograms(),
spectrograms(),
track_freq_contour()
{
# load data
data(list = c("Phae.long1", "Phae.long2", "lbh_selec_table"))
writeWave(Phae.long2, file.path(tempdir(), "Phae.long2.wav")) # save sound files
writeWave(Phae.long1, file.path(tempdir(), "Phae.long1.wav"))
# dominant frequency
freq_DTW(lbh_selec_table,
length.out = 30, flim = c(1, 12), bp = c(2, 9),
wl = 300, path = tempdir()
)
# fundamental frequency
freq_DTW(lbh_selec_table,
type = "fundamental", length.out = 30, flim = c(1, 12),
bp = c(2, 9), wl = 300, path = tempdir()
)
}
Run the code above in your browser using DataLab