seltailor
produces an interactive spectrographic view (similar to manualoc
) in
which the start and end times of acoustic signals listed in a data frame can be adjusted.
seltailor(X = NULL, wl = 512, flim = c(0,22), wn = "hanning", mar = 0.5, osci = FALSE, pal = reverse.gray.colors.2, ovlp = 70, auto.next = FALSE, pause = 1, comments = TRUE, path = NULL, frange = FALSE)
seltailor
or autodetec
can
be used as the input data frame. Other data frames can be used as input, but must have at least the 4 columns mentioned above. Required. Notice that, if an output file ("seltailor_output.csv") is found in the working directory it will be given priority over an input data frame.spectro
.
Default is c(0,22).ftwindow
for more options.TRUE
adds a oscillogram whenever the spectrograms are produced
with higher resolution (see seltime). Default is FALSE
.
The external program must be closed before resuming analysis. Default is NULL
.spectro
. Default is reverse.gray.colors.2. See Details.spectro
. Default is 70.TRUE
.NULL
(default) then the current working directory is used.manualoc
)
in which users can select a new start and end of a vocalization unit (e.g. elements)
by clicking at the end and then at the start of the signal (in any order). In addition, 2
"buttons" are provided at the upper right side of the spectrogram that
allow to stop the analysis ("Stop") or go to the next sound file ("next sel"). When a unit
has been selected, the function plots red dotted lines in the start and end of the
selection in the spectrogram. The lines "disappear" when a new selections is made.
Only the last selection is kept for each selection that is adjusted.
The function produces a .csv file (seltailor_output.csv) with the same information than the input
data frame, except for the new time coordinates, plus a new column (X$tailored) indicating if the selection
has been tailored. The file is saved in the working directory and is updated every time the user
moves into the next sound file (next sel "button") or stop the process
(Stop "button"). If no selection (by clicking on the 'next' buttom) the
original time coordinates are kept. When resuming the process (after "stop" and re-running
the function in the same working directory), the function will continue working on the
selections that have not been analyzed.
Windows length (wl) controls the temporal and frequency precision of the spectrogram.
A high "wl" value increases the frequency resolution but reduces the temporal resolution, and vice versa. Any
color palette that comes with the seewave package can be used: temp.colors,
reverse.gray.colors.1, reverse.gray.colors.2, reverse.heat.colors, reverse.terrain.colors,
reverse.topo.colors, reverse.cm.colors, heat.colors, terrain.colors, topo.colors,
cm.colors. Note that, unlike manualoc
, you cannot zoom in the spectrogram seltailor
.
The zoom can be adjusted by setting the mar
argument.
manualoc
## Not run:
# #First create empty folder
# setwd(tempdir())
#
# data(list = c("Phae.long1", "Phae.long2", "Phae.long3", "Phae.long4", "selec.table"))
# writeWave(Phae.long1,"Phae.long1.wav")
# writeWave(Phae.long2,"Phae.long2.wav")
# writeWave(Phae.long3,"Phae.long3.wav")
# writeWave(Phae.long4,"Phae.long4.wav")
#
# seltailor(X = selec.table, flim = c(1,12), wl = 300, auto.next = TRUE)
#
# # Read output .csv file
# seltailor.df <- read.csv("seltailor_output.csv")
# seltailor.df
#
# # check this directory for .csv file after stopping function
# getwd()
# ## End(Not run)
Run the code above in your browser using DataLab