static void audioGraph_FFTCalculated(object sender, SampleProcessor.FftEventArgs e) { var gainer = new MFCCGainer(26, e.Result.Count, e.SampleRate); buffer.Add(gainer.getCoefficents(e.Result)); var notes = classifier.getNotes(e.Result); //Console.WriteLine(String.Join(", ", notes.Distinct())); if (buffer.Count == 15) { if (!isTest) { Console.WriteLine("Current ErrorSum: {0}", network.errorSum); var results = network.Teach(buffer, MarkedValues[currentInstrument]); for (int i = 0; i < results.Count; i++) { Console.WriteLine("{0} - {1:00.00}%", instruments[i], results[i] * 100); } } else { var results = network.GetResults(buffer); for (int i = 0; i < results.Count; i++) { if (results[i] > 0.5) { testMap[currentInstrument].Add(instruments[i]); //Console.WriteLine("{0} - {1}", currentInstrument, instruments[i]); } } } buffer.RemoveAt(0); } }
static void audioGraph_FFTCalculated(object sender, SampleProcessor.FftEventArgs e) { var gainer = new MFCCGainer(26, e.Result.Count, e.SampleRate); buffer.Add(gainer.getCoefficents(e.Result)); if (buffer.Count == 10) { var result = network.GetResults(buffer); for (int i = 0; i < result.Count; i++) { if (result[i] > 0.4) { Console.WriteLine("{0} - {1}%", instruments[i], result[i] * 100); } } buffer.Clear(); } }