Package 'tdsc'

Convert a coding matrix to a distributed matrix

Description

This function converts a coding matrix of any kind into a distributed matrix as described in Farr (2007).

Usage

c2dmatrix(t, sf = 100)

Arguments

t	A tdsc object or a matrix
sf	The scaling factor

References

Farr (2007) “Automated Bioacoustic Identification of Statutory Quarantined Insect Pests”. PhD thesis. University of York.

Examples

c2dmatrix(as.matrix(c(1,2,3,4), nrow=2))

---

Coding Matrix from Chesmore (2001)

Description

Coding matrix used for Orthoptera.

Usage

data(chesmore2001)

Format

Matrix

Source

QTL Archive

References

Chesmore, E David (2001). “Application of time domain signal coding and artificial neural networks to passive acoustical identification of animals”. In: Applied Acoustics 62.12, pp. 1359–1374.

Examples

library(tuneR)
wave <- readWave(system.file("extdata", "1.wav", package="tdsc"))
data(chesmore2001)
t <- tdsc(wave, coding_matrix=chesmore2001)

---

Empty Bands Discovery

Description

Identifies unused codes across multiple S-matrices. Unused bands can be used to reduce the codebook as in Stammers (2011).

Usage

emptyBands(...)

Arguments

...	Two or more TDSC objects

References

Stammers (2011) “Audio Event Classification for Urban Soundscape Analysis”. PhD thesis. University of York.

Examples

library(tuneR)
wave <- readWave(system.file("extdata", "1.wav", package="tdsc"))
t <- tdsc(wave)
emptyBands(t,t)

---

Coding Matrix from Farr (2007)

Description

Coding matrix used by Farr (2007).

Usage

data(farr2007)

Format

Matrix

References

Farr (2007) “Automated Bioacoustic Identification of Statutory Quarantined Insect Pests”. PhD thesis. University of Hull.

Examples

library(tuneR)
wave <- readWave(system.file("extdata", "1.wav", package="tdsc"))
data(farr2007)
t <- tdsc(wave, coding_matrix=farr2007)

---

Following Codes

Description

Identifies sequences of codes that follow each other from time domain signal analysis, and optionally plots them as a Sankey diagram.

Usage

followingCodes(
  tdsc,
  depth = 2,
  min_code = 0,
  max_code = 10,
  colourCode = 1,
  plot = F,
  ...
)

Arguments

tdsc	A TDSC object
depth	The length of the sequence of codes to search for
min_code	The minimum value of code to include in sequence
max_code	The maximum value of code to include in sequence
colourCode	If plot is alluvial, colour all codes following this code
plot	If "alluvial" plots the found sequences in a river plot
...	Arguments to pass to the plotting function

Examples

## Not run: 
library(tuneR)
wave <- readWave(system.file("extdata", "1.wav", package="tdsc"))
t <- tdsc(wave)
followingCodes(t)
followingCodes(t, colourCode=2,plot="alluvial")

## End(Not run)

---

Normalise an A Matrix

Description

Normalises the A Matrix of a tdsc object either by scale or by codewords.

Usage

normalise.a.matrix(td, method = "scale")

Arguments

td	A tdsc object
method	Either scale (default) or codewords

Examples

library(tuneR)
wave <- readWave(system.file("extdata", "1.wav", package="tdsc"))
t <- tdsc(wave)
t <- normalise.a.matrix(t)
t <- normalise.a.matrix(t, method="codewords")

---

Visualise sampled waveforms

Description

Function to generate images of sampled waveforms with shapes analysed by Time Domain Signal Coding.

Usage

sample_waveform(
  samples = 3,
  fig_max_samples = NULL,
  start_zero = TRUE,
  invert = FALSE,
  tdsc_shapes = FALSE,
  limit_y = TRUE,
  ...
)

Arguments

samples	The number of samples
fig_max_samples	When constructing multiple figures this parameter can be used to ensure the plots are of the same size and are aligned
start_zero	If TRUE the waveform starts at zero, if FALSE the zero crossings are between samples
invert	If TRUE the shapes are positive with positive minima, if FALSE shapes are negative with negative maxima
tdsc_shapes	If TRUE the shapes correspond to TDSC shapes, if FALSE they resemble sampled sine waves
limit_y	If TRUE the shape fills the plot, if FALSE the complete range of the y axis is plotted (-1 to 1).
...	Further arguments to pass to plot.

Examples

sample_waveform()

---

Time Domain Signal Coding

Description

Performs Time Domain Signal Coding on a Wave object calculating the S-matrix and A-matrix.

Usage

tdsc(wave, lag = 1L, coding_matrix = NULL, plot = FALSE, max_D = 25L)

Arguments

wave	A Wave object
lag	The lag used to create the A-matrix
coding_matrix	A matrix used to code the Duration-Shape pairs
plot	If TRUE plots the workings of the coding algorithm
max_D	The maximum Duration to code

Examples

library(tuneR)
wave <- readWave(system.file("extdata", "1.wav", package="tdsc"))
t <- tdsc(wave)
t <- tdsc(wave, lag=2, max_D=10)

---

Plot the A matrix or S Matrix

Description

Plot the A matrix or S Matrix from a tdsc analysis.

Usage

tdsc_plot(td, plotter = "persp", ...)

Arguments

td	A TDSC object
plotter	Function used to plot the A matrix (persp or perp3D) or S matrix (hist)
...	Parameters to pass to plotting function

Examples

## Not run: 
tdsc.plot(td)
tdsc.plot(td, plotter="persp3D")
tdsc.plot(td, plotter="hist")

## End(Not run)

---

An S4 class to hold results from TDSC

Description

An S4 class to hold results from TDSC

Slots

raw

Two column vector of Durations and Shapes

positive

Identify non-negative sections

codelist

Vector of sequential epoch codings

b_matrix

The basic matrix

c_matrix

The coding matrix

s_matrix

The S-matrix

a_matrix

The A-matrix

sample_count

The number of samples in the waveform

epoch_count

The number of identified epochs

stdsc

Statistical TDSC feature vector

---