static void ClassifyAndDump(short[] signal, SignalClassifier classifier, string folder, string fileName)
{
var result = classifier.Classify(signal).Select(ConvertClassMark).ToArray();
var dump = new WavFileInfo(DefaultSampleRate, result);
var filePath = string.Concat(folder, fileName, "_result", ".wav");
dump.Write(filePath);
}
private static void ClassifyAll(string folder, string[] fileNames, SignalClassifier classifier)
{
foreach (var fileName in fileNames)
{
var wavFile = WavFileInfo.Read(string.Concat(folder, fileName, ".wav"));
if (wavFile.SampleRate != DefaultSampleRate)
{
throw new Exception(string.Format("File {0} has non-default sample rate", fileName));
}
ClassifyAndDump(wavFile.Data, classifier, folder, fileName);
var markerFile = MarkerFileInfo.Read(string.Concat(folder, fileName, ".mrk"));
var expectedWavFile = new WavFileInfo(wavFile.SampleRate,
markerFile.Decompress(wavFile.Data.Length)
.Select(ConvertClassMark)
.ToArray());
expectedWavFile.Write(string.Concat(folder, fileName, "_expected", ".wav"));
}
}
private static SignalClassifier TrainClassifier(string folder, string[] trainingSet, IFeatureExtractor[] featureExtractors, out double learningError)
{
var classifier = new SignalClassifier(featureExtractors);
foreach (var fileName in trainingSet)
{
var wavFile = WavFileInfo.Read(string.Concat(folder, fileName, ".wav"));
var markerFile = MarkerFileInfo.Read(string.Concat(folder, fileName, ".mrk"));
classifier.AddTrainingSignal(wavFile.Data, markerFile.Decompress(wavFile.Data.Length));
}
classifier.Learn(out learningError);
return classifier;
}
private static double EvaluateClassifier(string folder, string[] testSet, SignalClassifier classifier)
{
var errors = 0;
var total = 0;
foreach (var fileName in testSet)
{
var wavFile = WavFileInfo.Read(string.Concat(folder, fileName, ".wav"));
var markerFile = MarkerFileInfo.Read(string.Concat(folder, fileName, ".mrk"));
var actual = classifier.Classify(wavFile.Data);
var expected = markerFile.Decompress(wavFile.Data.Length);
total += wavFile.Data.Length;
errors += GetErrorsCount(actual, expected);
}
return errors * 1.0 / total;
}