public static float[] DecodeUsingMplayerAndSox(string fileIn, int srate, int secondsToAnalyze)
{
lock (_locker) {
using (Process tosoxreadable = new Process())
{
fileIn = Regex.Replace(fileIn, "%20", " ");
DbgTimer t = new DbgTimer();
t.Start();
String curdir = System.Environment.CurrentDirectory;
Dbg.WriteLine("Decoding: " + fileIn);
String tempFile = System.IO.Path.GetTempFileName();
String soxreadablewav = tempFile + ".wav";
Dbg.WriteLine("Temporary wav file: " + soxreadablewav);
tosoxreadable.StartInfo.FileName = "./NativeLibraries\\mplayer\\mplayer.exe";
tosoxreadable.StartInfo.Arguments = " -quiet -vc null -vo null -ao pcm:fast:waveheader \"" + fileIn + "\" -ao pcm:file=\\\"" + soxreadablewav + "\\\"";
tosoxreadable.StartInfo.UseShellExecute = false;
tosoxreadable.StartInfo.RedirectStandardOutput = true;
tosoxreadable.StartInfo.RedirectStandardError = true;
tosoxreadable.Start();
tosoxreadable.WaitForExit();
int exitCode = tosoxreadable.ExitCode;
// 0 = succesfull
// 1 = partially succesful
// 2 = failed
if (exitCode != 0)
{
string standardError = tosoxreadable.StandardError.ReadToEnd();
Console.Out.WriteLine(standardError);
return(null);
}
#if DEBUG
string standardOutput = tosoxreadable.StandardOutput.ReadToEnd();
Console.Out.WriteLine(standardOutput);
#endif
float[] floatBuffer = null;
if (File.Exists(soxreadablewav))
{
floatBuffer = DecodeUsingSox(soxreadablewav, srate, secondsToAnalyze);
try
{
File.Delete(tempFile);
File.Delete(soxreadablewav);
}
catch (IOException io)
{
Console.WriteLine(io);
}
}
Dbg.WriteLine("Decoding Execution Time: " + t.Stop().TotalMilliseconds + " ms");
return(floatBuffer);
}
}
}
public static float[] DecodeUsingMplayer(string fileIn, int srate)
{
lock (_locker) {
using (Process towav = new Process())
{
fileIn = Regex.Replace(fileIn, "%20", " ");
DbgTimer t = new DbgTimer();
t.Start();
String curdir = System.Environment.CurrentDirectory;
Dbg.WriteLine("Decoding: " + fileIn);
String tempFile = System.IO.Path.GetTempFileName();
String wav = tempFile + ".wav";
Dbg.WriteLine("Temporary wav file: " + wav);
towav.StartInfo.FileName = "./NativeLibraries\\mplayer\\mplayer.exe";
towav.StartInfo.Arguments = " -quiet -ao pcm:fast:waveheader \"" + fileIn + "\" -format floatle -af resample=" + srate + ":0:2,pan=1:0.5:0.5 -channels 1 -vo null -vc null -ao pcm:file=\\\"" + wav + "\\\"";
towav.StartInfo.UseShellExecute = false;
towav.StartInfo.RedirectStandardOutput = true;
towav.StartInfo.RedirectStandardError = true;
towav.Start();
towav.WaitForExit();
int exitCode = towav.ExitCode;
// 0 = succesfull
// 1 = partially succesful
// 2 = failed
if (exitCode != 0)
{
string standardError = towav.StandardError.ReadToEnd();
Console.Out.WriteLine(standardError);
return(null);
}
#if DEBUG
string standardOutput = towav.StandardOutput.ReadToEnd();
Console.Out.WriteLine(standardOutput);
#endif
RiffRead riff = new RiffRead(wav);
riff.Process();
float[] floatBuffer = riff.SoundData[0];
try
{
File.Delete(tempFile);
//File.Delete(wav);
}
catch (IOException io)
{
Console.WriteLine(io);
}
Dbg.WriteLine("Decoding Execution Time: " + t.Stop().TotalMilliseconds + " ms");
return(floatBuffer);
}
}
}
public Matrix Apply(ref Matrix m)
{
DbgTimer t = new DbgTimer();
t.Start();
Matrix mel = new Matrix(filterWeights.rows, m.columns);
/*
* // Performance optimization of ...
* mel = filterWeights.Multiply(m);
* for (int i = 0; i < mel.rows; i++) {
* for (int j = 0; j < mel.columns; j++) {
* mel.d[i, j] = (mel.d[i, j] < 1.0f ? 0 : (float)(10.0 * Math.Log10(mel.d[i, j])));
* //mel.d[i, j] = (float)(10.0 * Math.Log10(mel.d[i, j]));
* }
* }
*/
int mc = m.columns;
int mr = m.rows;
int melcolumns = mel.columns;
int fwc = filterWeights.columns;
int fwr = filterWeights.rows;
unsafe
{
fixed(float *md = m.d, fwd = filterWeights.d, meld = mel.d)
{
for (int i = 0; i < mc; i++)
{
for (int k = 0; k < fwr; k++)
{
int idx = k * melcolumns + i;
int kfwc = k * fwc;
for (int j = 0; j < mr; j++)
{
meld[idx] += fwd[kfwc + j] * md[j * mc + i];
}
meld[idx] = (meld[idx] < 1.0f ?
0 : (float)(10.0 * Math.Log10(meld[idx])));
}
}
}
}
Matrix mfcc = dct.Multiply(mel);
Dbg.WriteLine("mfcc (MfccLessOptimized) Execution Time: " + t.Stop().TotalMilliseconds + " ms");
return(mfcc);
}
/// <summary>
/// Computes a Scms model from the MFCC representation of a song.
/// </summary>
/// <param name="mfcc">Comirva.Audio.Util.Maths.Matrix mfcc</param>
/// <returns></returns>
public static Scms GetScmsNoInverse(Comirva.Audio.Util.Maths.Matrix mfccs, string name)
{
DbgTimer t = new DbgTimer();
t.Start();
Comirva.Audio.Util.Maths.Matrix mean = mfccs.Mean(2);
#if DEBUG
if (Analyzer.DEBUG_INFO_VERBOSE)
{
if (Analyzer.DEBUG_OUTPUT_TEXT)
{
mean.WriteText(name + "_mean.txt");
}
mean.DrawMatrixGraph(name + "_mean.png");
}
#endif
// Covariance
Comirva.Audio.Util.Maths.Matrix covarMatrix = mfccs.Cov(mean);
#if DEBUG
if (Analyzer.DEBUG_INFO_VERBOSE)
{
if (Analyzer.DEBUG_OUTPUT_TEXT)
{
covarMatrix.WriteText(name + "_covariance.txt");
}
covarMatrix.DrawMatrixGraph(name + "_covariance.png");
}
#endif
Comirva.Audio.Util.Maths.Matrix covarMatrixInv = new Comirva.Audio.Util.Maths.Matrix(covarMatrix.Rows, covarMatrix.Columns);
// Store the Mean, Covariance, Inverse Covariance in an optimal format.
int dim = mean.Rows;
Scms s = new Scms(dim);
int l = 0;
for (int i = 0; i < dim; i++)
{
s.mean[i] = (float)mean.MatrixData[i][0];
for (int j = i; j < dim; j++)
{
s.cov[l] = (float)covarMatrix.MatrixData[i][j];
s.icov[l] = (float)covarMatrixInv.MatrixData[i][j];
l++;
}
}
Dbg.WriteLine("GetScmsNoInverse - Execution Time: {0} ms", t.Stop().TotalMilliseconds);
return(s);
}
public Matrix Apply(ref Matrix m)
{
DbgTimer t = new DbgTimer();
t.Start();
Matrix mel = new Matrix(filterWeights.rows, m.columns);
int mc = m.columns;
int melcolumns = mel.columns;
int fwc = filterWeights.columns;
int fwr = filterWeights.rows;
unsafe
{
fixed(float *md = m.d, fwd = filterWeights.d, meld = mel.d)
{
for (int i = 0; i < mc; i++)
{
for (int k = 0; k < fwr; k++)
{
int idx = k * melcolumns + i;
int kfwc = k * fwc;
// The filter weights matrix is mostly 0.
// So only multiply non-zero elements!
for (int j = fwFT[k, 0]; j < fwFT[k, 1]; j++)
{
meld[idx] += fwd[kfwc + j] * md[j * mc + i];
}
meld[idx] = (meld[idx] < 1.0f ?
0 : (float)(10.0 * Math.Log10(meld[idx])));
}
}
}
}
try {
Matrix mfcc = dct.Multiply(mel);
long stop = 0;
t.Stop(ref stop);
Dbg.WriteLine("Mirage - mfcc Execution Time: {0}ms", stop);
return(mfcc);
} catch (MatrixDimensionMismatchException) {
throw new MfccFailedException();
}
}
public static Scms Analyze(string file_path)
{
DbgTimer t = new DbgTimer();
t.Start();
Matrix stftdata = ad.Decode(file_path);
Matrix mfccdata = mfcc.Apply(ref stftdata);
Scms scms = Scms.GetScms(mfccdata);
long stop = 0;
t.Stop(ref stop);
Dbg.WriteLine("Mirage - Total Execution Time: {0}ms", stop);
return(scms);
}
// Computes a Scms model from the MFCC representation of a song.
public static Scms GetScms(Matrix mfcc)
{
DbgTimer t = new DbgTimer();
t.Start();
// Mean
Vector m = mfcc.Mean();
// Covariance
Matrix c = mfcc.Covariance(m);
// Inverse Covariance
Matrix ic;
try {
ic = c.Inverse();
} catch (MatrixSingularException) {
throw new ScmsImpossibleException();
}
// Store the Mean, Covariance, Inverse Covariance in an optimal format.
int dim = m.rows;
Scms s = new Scms(dim);
int l = 0;
for (int i = 0; i < dim; i++)
{
s.mean[i] = m.d[i, 0];
for (int j = i; j < dim; j++)
{
s.cov[l] = c.d[i, j];
s.icov[l] = ic.d[i, j];
l++;
}
}
long stop = 0;
t.Stop(ref stop);
Dbg.WriteLine("Mirage - scms created in: {0}ms", stop);
return(s);
}
/// <summary>
/// Apply the STFT on the audiodata
/// </summary>
/// <param name="audiodata">Audiodata to apply the STFT on</param>
/// <returns>A matrix with the result of the STFT</returns>
public Matrix Apply(float[] audiodata)
{
DbgTimer t = new DbgTimer();
t.Start();
// calculate how many hops (bands) we have using the current overlap (hopsize)
int hops = (audiodata.Length - winsize) / hopsize;
// Create a Matrix with "winsize" Rows and "hops" Columns
// Matrix[Row, Column]
Matrix stft = new Matrix(winsize / 2 + 1, hops);
for (int i = 0; i < hops; i++)
{
fft.ComputeMirageMatrixUsingFftw(ref stft, i, audiodata, i * hopsize);
}
Dbg.WriteLine("Stft (ComputeMirageMatrix) Execution Time: " + t.Stop().TotalMilliseconds + " ms");
return(stft);
}
public Matrix Decode(string file)
{
int frames = 0;
int size = 0;
int ret = 0;
IntPtr data = mirageaudio_decode(ma, file, ref frames, ref size, ref ret);
if (ret == -1)
{
throw new AudioDecoderErrorException();
}
else if (ret == -2)
{
throw new AudioDecoderCanceledException();
}
else if ((frames <= 0) || (size <= 0))
{
throw new AudioDecoderErrorException(); // No data
}
Dbg.WriteLine("Mirage - decoded frames={0},size={1}", frames, size);
// Sort the frames by total energy (frame selection)
float [] frameselection = new float[frames];
int [] framepos = new int[frames];
unsafe {
float *stft_unsafe = (float *)data;
for (int j = 0; j < frames; j++)
{
frameselection[j] = 0;
framepos[j] = j;
for (int i = 0; i < size; i++)
{
frameselection[j] += stft_unsafe[i * frames + j];
}
}
}
Array.Sort(frameselection, framepos);
// Save the high energy frames to the Matrix
int copyframes = frames / 2;
Matrix stft = new Matrix(size, copyframes);
unsafe {
float *stft_unsafe = (float *)data;
fixed(float *stftd = stft.d)
{
for (int j = 0; j < copyframes; j++)
{
for (int i = 0; i < size; i++)
{
stftd[i * copyframes + j] = stft_unsafe[i * frames + framepos[copyframes + j]];
}
}
}
}
return(stft);
}
public static float[] Decode(string fileIn, int srate, int secondsToAnalyze)
{
DbgTimer t = new DbgTimer();
t.Start();
float[] floatBuffer = null;
// check if file exists
if (fileIn != null && fileIn != "")
{
FileInfo fi = new FileInfo(fileIn);
if (!fi.Exists)
{
Console.Out.WriteLine("No file found {0}!", fileIn);
return(null);
}
}
// Try to use Un4Seen Bass
BassProxy bass = BassProxy.Instance;
double duration = bass.GetDurationInSeconds(fileIn);
if (duration > 0)
{
Dbg.WriteLine("Using BASS to decode the file ...");
// duration in seconds
if (duration > secondsToAnalyze)
{
// find segment to extract
double startSeconds = (duration / 2 - (secondsToAnalyze / 2));
if (startSeconds < 0)
{
startSeconds = 0;
}
floatBuffer = bass.ReadMonoFromFile(fileIn, srate, secondsToAnalyze * 1000, (int)(startSeconds * 1000));
// if this failes, the duration read from the tags was wrong or it is something wrong with the audio file
if (floatBuffer == null)
{
IOUtils.LogMessageToFile(Mir.WARNING_FILES_LOG, fileIn);
}
}
else
{
// return whole file
floatBuffer = bass.ReadMonoFromFile(fileIn, srate, 0, 0);
// if this failes, the duration read from the tags was wrong or it is something wrong with the audio file
if (floatBuffer == null)
{
IOUtils.LogMessageToFile(Mir.WARNING_FILES_LOG, fileIn);
}
}
}
// Bass failed reading or never even tried, so use another alternative
if (floatBuffer == null)
{
Dbg.WriteLine("Using MPlayer and SOX to decode the file ...");
fileIn = Regex.Replace(fileIn, "%20", " ");
floatBuffer = DecodeUsingMplayerAndSox(fileIn, srate, secondsToAnalyze);
}
return(floatBuffer);
}
public static float[] DecodeUsingSox(string fileIn, int srate, int secondsToAnalyze)
{
lock (_locker) {
using (Process toraw = new Process())
{
fileIn = Regex.Replace(fileIn, "%20", " ");
DbgTimer t = new DbgTimer();
t.Start();
String curdir = System.Environment.CurrentDirectory;
Dbg.WriteLine("Decoding: " + fileIn);
String tempFile = System.IO.Path.GetTempFileName();
String raw = tempFile + "_raw.wav";
Dbg.WriteLine("Temporary raw file: " + raw);
toraw.StartInfo.FileName = "./NativeLibraries\\sox\\sox.exe";
toraw.StartInfo.Arguments = " \"" + fileIn + "\" -r " + srate + " -e float -b 32 -G -t raw \"" + raw + "\" channels 1";
toraw.StartInfo.UseShellExecute = false;
toraw.StartInfo.RedirectStandardOutput = true;
toraw.StartInfo.RedirectStandardError = true;
toraw.Start();
toraw.WaitForExit();
int exitCode = toraw.ExitCode;
// 0 = succesfull
// 1 = partially succesful
// 2 = failed
if (exitCode != 0)
{
string standardError = toraw.StandardError.ReadToEnd();
Console.Out.WriteLine(standardError);
return(null);
}
#if DEBUG
string standardOutput = toraw.StandardOutput.ReadToEnd();
Console.Out.WriteLine(standardOutput);
#endif
float[] floatBuffer;
FileStream fs = null;
try {
FileInfo fi = new FileInfo(raw);
fs = fi.OpenRead();
int bytes = (int)fi.Length;
int samples = bytes / sizeof(float);
if ((samples * sizeof(float)) != bytes)
{
return(null);
}
// if the audio file is larger than seconds to analyze,
// find a proper section to exctract
if (bytes > secondsToAnalyze * srate * sizeof(float))
{
int seekto = (bytes / 2) - ((secondsToAnalyze / 2) * sizeof(float) * srate);
Dbg.WriteLine("Extracting section: seekto = " + seekto);
bytes = (secondsToAnalyze) * srate * sizeof(float);
fs.Seek((samples / 2 - (secondsToAnalyze / 2) * srate) * sizeof(float), SeekOrigin.Begin);
}
BinaryReader br = new BinaryReader(fs);
byte[] bytesBuffer = new byte[bytes];
br.Read(bytesBuffer, 0, bytesBuffer.Length);
int items = (int)bytes / sizeof(float);
floatBuffer = new float[items];
for (int i = 0; i < items; i++)
{
floatBuffer[i] = BitConverter.ToSingle(bytesBuffer, i * sizeof(float)); // * 65536.0f;
}
} catch (System.IO.FileNotFoundException) {
floatBuffer = null;
} finally {
if (fs != null)
{
fs.Close();
}
try
{
File.Delete(tempFile);
File.Delete(raw);
}
catch (IOException io)
{
Console.WriteLine(io);
}
Dbg.WriteLine("Decoding Execution Time: " + t.Stop().TotalMilliseconds + " ms");
}
return(floatBuffer);
}
}
}
private void GaussJordan(ref decimal [,] a, int n, ref decimal [,] b, int m)
{
int [] indxc = new int[n + 1];
int [] indxr = new int[n + 1];
int [] ipiv = new int[n + 1];
int i, icol = 0, irow = 0, j, k, l, ll;
decimal big, dum, pivinv, temp;
for (j = 1; j <= n; j++)
{
ipiv[j] = 0;
}
for (i = 1; i <= n; i++)
{
big = 0;
for (j = 1; j <= n; j++)
{
if (ipiv[j] != 1)
{
for (k = 1; k <= n; k++)
{
if (ipiv[k] == 0)
{
if (Math.Abs(a[j, k]) >= big)
{
big = Math.Abs(a[j, k]);
irow = j;
icol = k;
}
}
else if (ipiv[k] > 1)
{
Dbg.WriteLine("Mirage - Gauss/Jordan Singular Matrix (1)");
throw new MatrixSingularException();
}
}
}
}
ipiv[icol]++;
if (irow != icol)
{
for (l = 1; l <= n; l++)
{
temp = a[irow, l];
a[irow, l] = a[icol, l];
a[icol, l] = temp;
}
for (l = 1; l <= m; l++)
{
temp = b[irow, l];
b[irow, l] = b[icol, l];
b[icol, l] = temp;
}
}
indxr[i] = irow;
indxc[i] = icol;
if (a[icol, icol] == 0)
{
Dbg.WriteLine("Mirage - Gauss/Jordan Singular Matrix (2)");
throw new MatrixSingularException();
}
pivinv = 1 / a[icol, icol];
a[icol, icol] = 1;
for (l = 1; l <= n; l++)
{
a[icol, l] *= pivinv;
}
for (l = 1; l <= m; l++)
{
b[icol, l] *= pivinv;
}
for (ll = 1; ll <= n; ll++)
{
if (ll != icol)
{
dum = a[ll, icol];
a[ll, icol] = 0;
for (l = 1; l <= n; l++)
{
a[ll, l] -= a[icol, l] * dum;
}
for (l = 1; l <= m; l++)
{
b[ll, l] -= b[icol, l] * dum;
}
}
}
}
for (l = n; l >= 1; l--)
{
if (indxr[l] != indxc[l])
{
for (k = 1; k <= n; k++)
{
temp = a[k, indxr[l]];
a[k, indxr[l]] = a[k, indxc[l]];
a[k, indxc[l]] = temp;
}
}
}
}
/// <summary>
/// Computes a Scms model from the MFCC representation of a song.
/// </summary>
/// <param name="mfcc">Comirva.Audio.Util.Maths.Matrix mfcc</param>
/// <returns></returns>
public static Scms GetScms(Comirva.Audio.Util.Maths.Matrix mfccs, string name)
{
DbgTimer t = new DbgTimer();
t.Start();
Comirva.Audio.Util.Maths.Matrix mean = mfccs.Mean(2);
#if DEBUG
if (Analyzer.DEBUG_INFO_VERBOSE)
{
if (Analyzer.DEBUG_OUTPUT_TEXT)
{
mean.WriteText(name + "_mean.txt");
}
mean.DrawMatrixGraph(name + "_mean.png");
}
#endif
// Covariance
Comirva.Audio.Util.Maths.Matrix covarMatrix = mfccs.Cov(mean);
#if DEBUG
if (Analyzer.DEBUG_INFO_VERBOSE)
{
if (Analyzer.DEBUG_OUTPUT_TEXT)
{
covarMatrix.WriteText(name + "_covariance.txt");
}
covarMatrix.DrawMatrixGraph(name + "_covariance.png");
}
#endif
// Inverse Covariance
Comirva.Audio.Util.Maths.Matrix covarMatrixInv;
try {
covarMatrixInv = covarMatrix.InverseGausJordan();
} catch (Exception) {
Dbg.WriteLine("MatrixSingularException - Scms failed!");
return(null);
}
#if DEBUG
if (Analyzer.DEBUG_INFO_VERBOSE)
{
if (Analyzer.DEBUG_OUTPUT_TEXT)
{
covarMatrixInv.WriteAscii(name + "_inverse_covariance.ascii");
}
covarMatrixInv.DrawMatrixGraph(name + "_inverse_covariance.png");
}
#endif
// Store the Mean, Covariance, Inverse Covariance in an optimal format.
int dim = mean.Rows;
Scms s = new Scms(dim);
int l = 0;
for (int i = 0; i < dim; i++)
{
s.mean[i] = (float)mean.MatrixData[i][0];
for (int j = i; j < dim; j++)
{
s.cov[l] = (float)covarMatrix.MatrixData[i][j];
s.icov[l] = (float)covarMatrixInv.MatrixData[i][j];
l++;
}
}
Dbg.WriteLine("Compute Scms - Execution Time: {0} ms", t.Stop().TotalMilliseconds);
return(s);
}
/// <summary>
/// Computes a Scms model from the MFCC representation of a song.
/// </summary>
/// <param name="mfcc">Mirage.Matrix mfcc</param>
/// <returns></returns>
public static Scms GetScms(Matrix mfcc, string name)
{
DbgTimer t = new DbgTimer();
t.Start();
// Mean
Vector m = mfcc.Mean();
#if DEBUG
if (Analyzer.DEBUG_INFO_VERBOSE)
{
if (Analyzer.DEBUG_OUTPUT_TEXT)
{
m.WriteText(name + "_mean_orig.txt");
}
m.DrawMatrixGraph(name + "_mean_orig.png");
}
#endif
// Covariance
Matrix c = mfcc.Covariance(m);
#if DEBUG
if (Analyzer.DEBUG_INFO_VERBOSE)
{
if (Analyzer.DEBUG_OUTPUT_TEXT)
{
c.WriteText(name + "_covariance_orig.txt");
}
c.DrawMatrixGraph(name + "_covariance_orig.png");
}
#endif
// Inverse Covariance
Matrix ic;
try {
ic = c.Inverse();
} catch (MatrixSingularException) {
//throw new ScmsImpossibleException();
Dbg.WriteLine("MatrixSingularException - Scms failed!");
return(null);
}
#if DEBUG
if (Analyzer.DEBUG_INFO_VERBOSE)
{
if (Analyzer.DEBUG_OUTPUT_TEXT)
{
ic.WriteAscii(name + "_inverse_covariance_orig.txt");
}
ic.DrawMatrixGraph(name + "_inverse_covariance_orig.png");
}
#endif
// Store the Mean, Covariance, Inverse Covariance in an optimal format.
int dim = m.rows;
Scms s = new Scms(dim);
int l = 0;
for (int i = 0; i < dim; i++)
{
s.mean[i] = m.d[i, 0];
for (int j = i; j < dim; j++)
{
s.cov[l] = c.d[i, j];
s.icov[l] = ic.d[i, j];
l++;
}
}
Dbg.WriteLine("(Mirage) - scms created in: {0} ms", t.Stop().TotalMilliseconds);
return(s);
}
