信号処理

Signalクラス

http://accord-framework.net/docs/html/T_Accord_Audio_Signal.htm

// create an empty audio signal 
Signal signal = new Signal( channels, length, sampleRate, format );
Copy
float[,] data = 
{
    {  0.00f, 0.2f  },
    {  0.32f, 0.1f  },
    {  0.22f, 0.2f  },
    {  0.12f, 0.42f },
    { -0.12f, 0.1f  },
    { -0.22f, 0.2f  },
};

// or create an audio signal from an array of audio frames
Signal target = Signal.FromArray(data, sampleRate: 8000);

WaveFileAudioSourceクラス

http://accord-framework.net/docs/html/T_Accord_DirectSound_WaveFileAudioSource.htm

// Create the Wave file audio source
WaveFileAudioSource source = new WaveFileAudioSource("audiofile.wav");

// Specify the callback function which will be
// called once a sample is completely available
source.NewFrame += source_NewFrame;

// Start capturing
source.Start();

// ...

// The callback function should determine what
// should be done with the samples being caught
private void source_NewFrame(object sender, NewFrameEventArgs eventArgs)
{
    // Read current frame...
    Signal s = eventArgs.Signal;

    // Process/play/record it
    // ...
}

WaveDecoder Classクラス

Complex Signal Class

http://accord-framework.net/docs/html/T_Accord_Audio_ComplexSignal.htm

// Create complex audio signal
ComplexSignal complexSignal = ComplexSignal.FromSignal( signal );

// Do forward Fourier transformation
complexSignal.ForwardFourierTransform( );

// Generate spectrogram
complexSignal.ToBitmap(512,512);

However, if your signal is too lengthy, or if your signal is not yet in a power of two size, you can use a temporal window to slice your signal into smaller cuts, as shown below. In the example, an audio file is being read and its contents are being decoded and stored into a Signal object. Afterwards, an audio window is being used to cut the signal into smaller, power-of-two size signals which can then be transformed into the frequency (Fourier) domain. Copy

string fileName = "audio.wav";

WaveDecoder sourceDecoder = new WaveDecoder(fileName);

// Decode the file and store into a signal
Signal sourceSignal = sourceDecoder.Decode();

// Create Hamming window so that signal will fit into power of 2:           
RaisedCosineWindow window = RaisedCosineWindow.Hamming(1024);

// Splits the source signal by walking each 512 samples, then creating 
// a 1024 sample window. Note that this will result in overlapped windows.
Signal[] windows = sourceSignal.Split(window, 512);

// You might need to import Accord.Math in order to call this:
ComplexSignal[] complex = windows.Apply(ComplexSignal.FromSignal);

// Forward to the Fourier domain
complex.ForwardFourierTransform();