Class BirdClefExperiment1

This class is experimental work on the Bird50 dataset provided by Herve Glotin. The work was started in Toulon (February 2016) and continued after my return in March 2016. The bird50 dataset is a randomly selected subset of the 2014 BirdClef 500 data set. The 2014 competition was won by Dan Stowell (QMUL). The 2015 competition containing 999 bird call recordings was won by Mario Lessek (Berlin)

This class prepares species representations of bird calls by summing or averaging the instance representations. The representation consists of the concatenation of a set of spectra. There are currently five spectra derived from the spectra indices: SPT, RHZ, RVT, RPS and RNG. Each spectrum can be reduced from 256 values to say 160 by max pooling the top end of the spctrum. NOTE: MEL-scale DOES NOT work for birds because the dominant activity for birds is around 2-4 kHz. The MEL-scale effectively obliterates this band of the spectrum.

Inheritance

Object

BirdClefExperiment1

Inherited Members

Object.Equals(Object)

Object.Equals(Object, Object)

Object.GetHashCode()

Object.GetType()

Object.MemberwiseClone()

Object.ReferenceEquals(Object, Object)

Object.ToString()

Namespace: AudioAnalysisTools

Assembly: AudioAnalysisTools.dll

Syntax

public static class BirdClefExperiment1

Methods

| Improve this Doc View Source

CalculateAccuracy(BirdClefExperiment1.Arguments, BirdClefExperiment1.Output)

Produce a CONFUSION MATRIX and a RANK ORDER MATRIX.

Declaration

public static void CalculateAccuracy(BirdClefExperiment1.Arguments arguments, BirdClefExperiment1.Output output)

Parameters

Type	Name	Description
BirdClefExperiment1.Arguments	arguments
BirdClefExperiment1.Output	output

| Improve this Doc View Source

CalculateSimilarityScores(BirdClefExperiment1.Arguments, BirdClefExperiment1.Output)

This done using Cosine similarity. Could also use Euclidian distance.

Declaration

public static void CalculateSimilarityScores(BirdClefExperiment1.Arguments arguments, BirdClefExperiment1.Output output)

Parameters

Type	Name	Description
BirdClefExperiment1.Arguments	arguments
BirdClefExperiment1.Output	output

| Improve this Doc View Source

ConstructWekaDatasets(BirdClefExperiment1.Arguments, BirdClefExperiment1.Output)

Construct datasets for WEKA machine learning.

Declaration

public static void ConstructWekaDatasets(BirdClefExperiment1.Arguments arguments, BirdClefExperiment1.Output output)

Parameters

Type	Name	Description
BirdClefExperiment1.Arguments	arguments
BirdClefExperiment1.Output	output

| Improve this Doc View Source

DrawSpeciesImages(BirdClefExperiment1.Arguments, BirdClefExperiment1.Output)

Declaration

public static void DrawSpeciesImages(BirdClefExperiment1.Arguments arguments, BirdClefExperiment1.Output output)

Parameters

Type	Name	Description
BirdClefExperiment1.Arguments	arguments
BirdClefExperiment1.Output	output

| Improve this Doc View Source

Execute(BirdClefExperiment1.Arguments)

Declaration

public static void Execute(BirdClefExperiment1.Arguments arguments)

Parameters

Type	Name	Description
BirdClefExperiment1.Arguments	arguments

| Improve this Doc View Source

ExoticMaxPoolingMatrixColumns(Double[,], Int32)

Declaration

public static double[, ] ExoticMaxPoolingMatrixColumns(double[, ] matrix, int reducedColCount)

Parameters

Type	Name	Description
Double[,]	matrix
Int32	reducedColCount

Returns

Type	Description
Double[,]

| Improve this Doc View Source

GetInstanceRepresentations(BirdClefExperiment1.Arguments)

Declaration

public static BirdClefExperiment1.Output GetInstanceRepresentations(BirdClefExperiment1.Arguments arguments)

Parameters

Type	Name	Description
BirdClefExperiment1.Arguments	arguments

Returns

Type	Description
BirdClefExperiment1.Output

| Improve this Doc View Source

GetSpeciesRepresentations(BirdClefExperiment1.Arguments, BirdClefExperiment1.Output)

Declaration

public static void GetSpeciesRepresentations(BirdClefExperiment1.Arguments arguments, BirdClefExperiment1.Output output)

Parameters

Type	Name	Description
BirdClefExperiment1.Arguments	arguments
BirdClefExperiment1.Output	output

| Improve this Doc View Source

MaxPoolingLimited(Double[], Int32, Int32, Int32, Int32)

reduces the dimensionality of a vector by max pooling. Used specifically for representation of spectral frames in Herve Glotin work.

Declaration

public static double[] MaxPoolingLimited(double[] vector, int startBin, int maxOf2Bin, int maxOf3Bin, int endBin)

Parameters

Type	Name	Description
Double[]	vector
Int32	startBin
Int32	maxOf2Bin
Int32	maxOf3Bin
Int32	endBin

Returns

Type	Description
Double[]

| Improve this Doc View Source

MaxPoolingLimited(Double[,], Int32, Int32, Int32, Int32, Int32)

Declaration

public static double[, ] MaxPoolingLimited(double[, ] M, int startBin, int maxOf2Bin, int maxOf3Bin, int endBin, int reducedBinCount)

Parameters

Type	Name	Description
Double[,]	M
Int32	startBin
Int32	maxOf2Bin
Int32	maxOf3Bin
Int32	endBin
Int32	reducedBinCount

Returns

Type	Description
Double[,]

| Improve this Doc View Source

MaxPoolMatrixColumns(Double[,], Int32)

Declaration

public static double[, ] MaxPoolMatrixColumns(double[, ] matrix, int reducedColCount)

Parameters

Type	Name	Description
Double[,]	matrix
Int32	reducedColCount

Returns

Type	Description
Double[,]

| Improve this Doc View Source

MaxPoolMatrixColumnsByFactor(Double[,], Int32)

Declaration

public static double[, ] MaxPoolMatrixColumnsByFactor(double[, ] matrix, int factor)

Parameters

Type	Name	Description
Double[,]	matrix
Int32	factor

Returns

Type	Description
Double[,]

| Improve this Doc View Source

Normalise(BirdClefExperiment1.Arguments, BirdClefExperiment1.Output)

Normalisation and Concatentation of spectra: can be done in three ways ie (i) Unit length (ii) Unit Area (iii) Unit bounds i.e. 0,1.

Declaration

public static void Normalise(BirdClefExperiment1.Arguments arguments, BirdClefExperiment1.Output output)

Parameters

Type	Name	Description
BirdClefExperiment1.Arguments	arguments
BirdClefExperiment1.Output	output

| Improve this Doc View Source

NormaliseVector(Double[], Double[])

Normalises the parts of a concatenated vector separately. Finally does a unit length norm in prepration for a dot product to give cosine similairty.

Declaration

public static double[] NormaliseVector(double[] ipVector, double[] weights)

Parameters

Type	Name	Description
Double[]	ipVector
Double[]	weights

Returns

Type	Description
Double[]

| Improve this Doc View Source

ReadGlotinsSpeciesLabelFile(String, Int32, out String[], out Int32[])

Declaration

public static void ReadGlotinsSpeciesLabelFile(string speciesLabelsFile, int count, out string[] fileID, out int[] speciesID)

Parameters

Type	Name	Description
String	speciesLabelsFile
Int32	count
String[]	fileID
Int32[]	speciesID

| Improve this Doc View Source

ReduceMatrixColumns(Double[,], Int32, Int32)

Declaration

public static double[, ] ReduceMatrixColumns(double[, ] matrix, int minBin, int maxBin)

Parameters

Type	Name	Description
Double[,]	matrix
Int32	minBin
Int32	maxBin

Returns

Type	Description
Double[,]

| Improve this Doc View Source

WriteDataTODOTODOTODOTODO()

Declaration

public static void WriteDataTODOTODOTODOTODO()