[src] Adding some more feature extraction options (needed by some use…

…rs..) (kaldi-asr#3724)

src/feat/feature-spectrogram.cc

@@ -62,6 +62,11 @@ void SpectrogramComputer::Compute(BaseFloat signal_raw_log_energy,
   else  // An alternative algorithm that works for non-powers-of-two
     RealFft(signal_frame, true);
 
+  if (opts_.return_raw_fft) {
+    feature->CopyFromVec(*signal_frame);
+    return;
+  }
+
   // Convert the FFT into a power spectrum.
   ComputePowerSpectrum(signal_frame);
   SubVector<BaseFloat> power_spectrum(*signal_frame,

src/feat/feature-spectrogram.h

@@ -39,10 +39,14 @@ struct SpectrogramOptions {
   FrameExtractionOptions frame_opts;
   BaseFloat energy_floor;
   bool raw_energy;  // If true, compute energy before preemphasis and windowing
+  bool return_raw_fft; // If true, return the raw FFT spectrum
+                       // Note that in that case the Dim() will return double
+                       // the expected dimension (because of the complex domain of it)
 
   SpectrogramOptions() :
     energy_floor(0.0),
-    raw_energy(true) {}
+    raw_energy(true),
+    return_raw_fft(false) {}
 
   void Register(OptionsItf *opts) {
     frame_opts.Register(opts);
@@ -54,6 +58,8 @@ struct SpectrogramOptions {
                    "std::numeric_limits<float>::epsilon().");
     opts->Register("raw-energy", &raw_energy,
                    "If true, compute energy before preemphasis and windowing");
+    opts->Register("return-raw-fft", &return_raw_fft,
+                   "If true, return raw FFT complex numbers instead of log magnitudes");
   }
 };
 
@@ -68,7 +74,13 @@ class SpectrogramComputer {
     return opts_.frame_opts;
   }
 
-  int32 Dim() const { return opts_.frame_opts.PaddedWindowSize() / 2 + 1; }
+  int32 Dim() const {
+    if (opts_.return_raw_fft) {
+      return opts_.frame_opts.PaddedWindowSize();
+    } else {
+      return opts_.frame_opts.PaddedWindowSize() / 2 + 1;
+    }
+  }
 
   bool NeedRawLogEnergy() const { return opts_.raw_energy; }
 

src/feat/feature-window.cc

@@ -115,6 +115,10 @@ FeatureWindowFunction::FeatureWindowFunction(const FrameExtractionOptions &opts)
     double i_fl = static_cast<double>(i);
     if (opts.window_type == "hanning") {
       window(i) = 0.5  - 0.5*cos(a * i_fl);
+    } else if (opts.window_type == "sine") {
+      // when you are checking ws wikipedia, please
+      // note that 0.5 * a = M_PI/(frame_length-1)
+      window(i) = sin(0.5 * a * i_fl);
     } else if (opts.window_type == "hamming") {
       window(i) = 0.54 - 0.46*cos(a * i_fl);
     } else if (opts.window_type == "povey") {  // like hamming but goes to zero at edges.

src/feat/feature-window.h

@@ -40,7 +40,7 @@ struct FrameExtractionOptions {
   BaseFloat preemph_coeff;  // Preemphasis coefficient.
   bool remove_dc_offset;  // Subtract mean of wave before FFT.
   std::string window_type;  // e.g. Hamming window
-  // May be "hamming", "rectangular", "povey", "hanning", "blackman"
+  // May be "hamming", "rectangular", "povey", "hanning", "sine", "blackman"
   // "povey" is a window I made to be similar to Hamming but to go to zero at the
   // edges, it's pow((0.5 - 0.5*cos(n/N*2*pi)), 0.85)
   // I just don't think the Hamming window makes sense as a windowing function.
@@ -81,7 +81,7 @@ struct FrameExtractionOptions {
                    "option, e.g. to 1.0 or 0.1");
     opts->Register("window-type", &window_type, "Type of window "
                    "(\"hamming\"|\"hanning\"|\"povey\"|\"rectangular\""
-                   "|\"blackmann\")");
+                   "|\"sine\"|\"blackmann\")");
     opts->Register("blackman-coeff", &blackman_coeff,
                    "Constant coefficient for generalized Blackman window.");
     opts->Register("round-to-power-of-two", &round_to_power_of_two,