public List<bool[]> CreateFingerprintsFromAudioSamples(float[] samples, WorkUnitParameterObject param, out double[][] logSpectrogram, out List<double[][]> spectralImages)
{
IFingerprintingConfiguration configuration = param.FingerprintingConfiguration;
AudioServiceConfiguration audioServiceConfiguration = new AudioServiceConfiguration
{
LogBins = configuration.LogBins,
LogBase = configuration.LogBase,
MaxFrequency = configuration.MaxFrequency,
MinFrequency = configuration.MinFrequency,
Overlap = configuration.Overlap,
SampleRate = configuration.SampleRate,
WindowSize = configuration.WindowSize,
NormalizeSignal = configuration.NormalizeSignal,
UseDynamicLogBase = configuration.UseDynamicLogBase
};
// store the log spectrogram in the out variable
logSpectrogram = AudioService.CreateLogSpectrogram(
samples, configuration.WindowFunction, audioServiceConfiguration);
return this.CreateFingerprintsFromLogSpectrum(
logSpectrogram,
configuration.Stride,
configuration.FingerprintLength,
configuration.Overlap,
configuration.TopWavelets,
out spectralImages);
}
public float[][] CreateLogSpectrogram(
float[] samples, IWindowFunction windowFunction, AudioServiceConfiguration configuration)
{
NormalizeInPlace(samples);
int width = (samples.Length - configuration.WdftSize) / configuration.Overlap; /*width of the image*/
float[][] frames = new float[width][];
float[] complexSignal = new float[2 * configuration.WdftSize]; /*even - Re, odd - Img*/
double[] window = windowFunction.GetWindow(configuration.WdftSize);
int[] logFrequenciesIndexes = GenerateLogFrequencies(configuration);
for (int i = 0; i < width; i++)
{
// take 371 ms each 11.6 ms (2048 samples each 64 samples)
for (int j = 0; j < configuration.WdftSize /*2048*/; j++)
{
complexSignal[(2 * j)] = (float)(window[j] * samples[(i * configuration.Overlap) + j]);
/*Weight by Hann Window*/
complexSignal[(2 * j) + 1] = 0;
}
// FFT transform for gathering the spectrum
Fourier.FFT(complexSignal, configuration.WdftSize, FourierDirection.Forward);
frames[i] = ExtractLogBins(complexSignal, logFrequenciesIndexes, configuration.LogBins);
}
return frames;
}
private List<bool[]> CreateFingerprintsFromAudioSamples(float[] samples, WorkUnitParameterObject param)
{
IFingerprintingConfiguration configuration = param.FingerprintingConfiguration;
AudioServiceConfiguration audioServiceConfiguration = new AudioServiceConfiguration
{
LogBins = configuration.LogBins,
LogBase = configuration.LogBase,
MaxFrequency = configuration.MaxFrequency,
MinFrequency = configuration.MinFrequency,
Overlap = configuration.Overlap,
SampleRate = configuration.SampleRate,
WdftSize = configuration.WdftSize
};
float[][] spectrum = audioService.CreateLogSpectrogram(
samples, configuration.WindowFunction, audioServiceConfiguration);
return CreateFingerprintsFromSpectrum(
spectrum,
configuration.Stride,
configuration.FingerprintLength,
configuration.Overlap,
configuration.LogBins,
configuration.TopWavelets);
}
public double[][] CreateLogSpectrogram(
float[] samples, IWindowFunction windowFunction, AudioServiceConfiguration configuration)
{
// Explode samples to the range of 16 bit shorts (–32,768 to 32,767)
// Matlab multiplies with 2^15 (32768)
// e.g. if( max(abs(speech))<=1 ), speech = speech * 2^15; end;
//MathUtils.Multiply(ref samples, Analyzer.AUDIO_MULTIPLIER); // 65536
if (configuration.NormalizeSignal)
{
NormalizeInPlace(samples);
}
int width = (samples.Length - configuration.WdftSize) / configuration.Overlap; /*width of the image*/
double[][] frames = new double[width][];
int[] logFrequenciesIndexes = GenerateLogFrequencies(configuration);
//double[] window = windowFunction.GetWindow(configuration.WdftSize);
double[] window = windowFunction.GetWindow();
for (int i = 0; i < width; i++)
{
double[] complexSignal = new double[2 * configuration.WdftSize]; /*even - Re, odd - Img, thats how Exocortex works*/
// take 371 ms each 11.6 ms (2048 samples each 64 samples)
for (int j = 0; j < configuration.WdftSize; j++)
{
// Weight by Hann Window
complexSignal[2 * j] = window[j] * samples[(i * configuration.Overlap) + j];
// need to clear out as fft modifies buffer (phase)
complexSignal[(2 * j) + 1] = 0;
}
lomonFFT.TableFFT(complexSignal, true);
frames[i] = ExtractLogBins(complexSignal, logFrequenciesIndexes, configuration.LogBins);
}
return frames;
}
public double[][] CreateLogSpectrogram(string pathToFile, IWindowFunction windowFunction, AudioServiceConfiguration configuration)
{
float[] samples = ReadMonoFromFile(pathToFile, configuration.SampleRate, 0, 0);
return CreateLogSpectrogram(samples, windowFunction, configuration);
}
private int[] GenerateStaticLogFrequencies(AudioServiceConfiguration configuration)
{
double logMin = Math.Log(configuration.MinFrequency, configuration.LogBase);
double logMax = Math.Log(configuration.MaxFrequency, configuration.LogBase);
double delta = (logMax - logMin) / configuration.LogBins;
int[] indexes = new int[configuration.LogBins + 1];
double accDelta = 0;
for (int i = 0; i <= configuration.LogBins /*32 octaves*/; ++i)
{
double freq = (double)Math.Pow(configuration.LogBase, logMin + accDelta);
accDelta += delta; // accDelta = delta * i
/*Find the start index in array from which to start the summation*/
indexes[i] = FreqToIndex(freq, configuration.SampleRate, configuration.WdftSize);
}
return indexes;
}
private int[] GenerateLogFrequenciesDynamicBase(AudioServiceConfiguration configuration)
{
double logBase =
Math.Exp(
Math.Log((double)configuration.MaxFrequency / configuration.MinFrequency) / configuration.LogBins);
double mincoef = (double)configuration.WdftSize / configuration.SampleRate * configuration.MinFrequency;
int[] indexes = new int[configuration.LogBins + 1];
for (int j = 0; j < configuration.LogBins + 1; j++)
{
int start = (int)((Math.Pow(logBase, j) - 1.0) * mincoef);
int end = (int)((Math.Pow(logBase, j + 1.0f) - 1.0) * mincoef);
indexes[j] = start + (int)mincoef;
}
return indexes;
}
/// <summary>
/// Get logarithmically spaced indices
/// </summary>
/// <param name="configuration">
/// The configuration for log frequencies
/// </param>
/// <returns>
/// Log indexes
/// </returns>
private int[] GenerateLogFrequencies(AudioServiceConfiguration configuration)
{
if(configuration.UseDynamicLogBase)
{
return GenerateLogFrequenciesDynamicBase(configuration);
}
return GenerateStaticLogFrequencies(configuration);
}
public double[][] CreateLogSpectrogram(
float[] samples, IWindowFunction windowFunction, AudioServiceConfiguration configuration)
{
DbgTimer t = new DbgTimer();
t.Start ();
if (configuration.NormalizeSignal)
{
NormalizeInPlace(samples);
}
int width = (samples.Length - configuration.WindowSize) / configuration.Overlap; /*width of the image*/
double[][] frames = new double[width][];
int[] logFrequenciesIndexes = GenerateLogFrequencies(configuration);
double[] window = windowFunction.GetWindow();
for (int i = 0; i < width; i++)
{
double[] complexSignal = new double[2 * configuration.WindowSize]; /*even - Re, odd - Img, thats how Exocortex works*/
// take 371 ms each 11.6 ms (2048 samples each 64 samples, samplerate 5512)
// or 256 ms each 16 ms (8192 samples each 512 samples, samplerate 32000)
for (int j = 0; j < configuration.WindowSize; j++)
{
// Weight by Hann Window
complexSignal[2 * j] = window[j] * samples[(i * configuration.Overlap) + j];
// need to clear out as fft modifies buffer (phase)
complexSignal[(2 * j) + 1] = 0;
}
lomonFFT.TableFFT(complexSignal, true);
frames[i] = ExtractLogBins(complexSignal, logFrequenciesIndexes, configuration.LogBins);
}
Dbg.WriteLine ("Create Log Spectrogram - Execution Time: {0} ms", t.Stop().TotalMilliseconds);
return frames;
}
private static void TestSoundfingerprintingAlgorithm(string filename, string name)
{
// work config
WorkUnitParameterObject param = new WorkUnitParameterObject();
param.PathToAudioFile = filename;
param.StartAtMilliseconds = 0;
param.MillisecondsToProcess = 0;
param.FingerprintingConfiguration = fingerprintingConfig;
// Soundfingerprinting Service
FingerprintService fingerprintService = GetSoundfingerprintingService();
// Image Service
ImageService imageService = new ImageService(
fingerprintService.SpectrumService,
fingerprintService.WaveletService);
// Configuration
AudioServiceConfiguration audioServiceConfiguration = new AudioServiceConfiguration
{
LogBins = fingerprintingConfig.LogBins,
LogBase = fingerprintingConfig.LogBase,
MaxFrequency = fingerprintingConfig.MaxFrequency,
MinFrequency = fingerprintingConfig.MinFrequency,
Overlap = fingerprintingConfig.Overlap,
SampleRate = fingerprintingConfig.SampleRate,
WdftSize = fingerprintingConfig.WdftSize,
NormalizeSignal = fingerprintingConfig.NormalizeSignal,
UseDynamicLogBase = fingerprintingConfig.UseDynamicLogBase
};
double[][] spectrogram = fingerprintService.AudioService.CreateSpectrogram(filename, new Mirage.HannWindow(fingerprintingConfig.WdftSize), fingerprintingConfig.SampleRate, fingerprintingConfig.Overlap, fingerprintingConfig.WdftSize);
imageService.GetSpectrogramImage(spectrogram, 600, 400).Save("imageservice_" + name + "_specgram.png");
/*
Comirva.Audio.Util.Maths.Matrix stftdata = new Comirva.Audio.Util.Maths.Matrix(spectrogram).Transpose();
#if DEBUG
if (Analyzer.DEBUG_INFO_VERBOSE) {
if (DEBUG_OUTPUT_TEXT) {
stftdata.WriteAscii(name + "_stftdata2.ascii");
stftdata.WriteCSV(name + "_stftdata2.csv", ";");
}
// same as specgram(audio*32768, 2048, 44100, hanning(2048), 1024);
stftdata.DrawMatrixImageLogValues(name + "_specgram2.png", true);
// spec gram with log values for the y axis (frequency)
stftdata.DrawMatrixImageLogY(name + "_specgramlog2.png", SAMPLING_RATE, 20, SAMPLING_RATE/2, 120, WINDOW_SIZE);
}
#endif
*/
double[][] logSpectrogram = fingerprintService.AudioService.CreateLogSpectrogram(filename, new Mirage.HannWindow(fingerprintingConfig.WdftSize), audioServiceConfiguration);
imageService.GetLogSpectralImages(logSpectrogram, fingerprintingConfig.Stride, fingerprintingConfig.FingerprintLength, fingerprintingConfig.Overlap, 2).Save("imageservice_" + name + "_specgram_logimages.png");
Comirva.Audio.Util.Maths.Matrix stftdataLog = new Comirva.Audio.Util.Maths.Matrix(logSpectrogram).Transpose();
#if DEBUG
if (Analyzer.DEBUG_INFO_VERBOSE) {
if (DEBUG_OUTPUT_TEXT) {
stftdataLog.WriteAscii(name + "_stftdataLog.ascii");
stftdataLog.WriteCSV(name + "_stftdataLog.csv", ";");
}
// same as specgram(audio*32768, 2048, 44100, hanning(2048), 1024);
stftdataLog.DrawMatrixImageLogValues(name + "_stftdataLog.png", true);
}
#endif
// Get fingerprints
double[][] LogSpectrogram;
List<bool[]> fingerprints = fingerprintService.CreateFingerprintsFromAudioFile(param, out LogSpectrogram);
int width = fingerprintingConfig.FingerprintLength;
int height = fingerprintingConfig.LogBins;
imageService.GetImageForFingerprints(fingerprints, width, height, 2).Save("imageservice_" + name + "_fingerprints.png");
/*
IPermutations permutations = new LocalPermutations("Soundfingerprinting\\perms.csv", ",");
Soundfingerprinting.DuplicatesDetector.DataAccess.Repository repository = new Soundfingerprinting.DuplicatesDetector.DataAccess.Repository(permutations);
// Define track
Soundfingerprinting.DuplicatesDetector.Model.Track track
= new Soundfingerprinting.DuplicatesDetector.Model.Track {
Title = name,
Path = filename
};
// Get the HashSignatures
List<Soundfingerprinting.DuplicatesDetector.Model.HashSignature> signatures = repository.GetSignatures(fingerprints, track, 25, 4);
return signatures;
*/
}
