shift-feats.cc File Reference

#include "base/kaldi-common.h"
#include "util/common-utils.h"
#include "matrix/kaldi-matrix.h"

Include dependency graph for shift-feats.cc:

Go to the source code of this file.

Namespaces
	kaldi
	This code computes Goodness of Pronunciation (GOP) and extracts phone-level pronunciation feature for mispronunciations detection tasks, the reference:

Functions
void	ShiftFeatureMatrix (const Matrix< BaseFloat > &src, int32 shift, Matrix< BaseFloat > *rearranged)
int	main (int argc, char *argv[])

Function Documentation

main()

int main	(	int	argc,
		char *	argv[]
	)

Definition at line 37 of file shift-feats.cc.

References kaldi::ClassifyRspecifier(), ParseOptions::GetArg(), KALDI_ERR, KALDI_LOG, KALDI_WARN, kaldi::kNoRspecifier, ParseOptions::NumArgs(), MatrixBase< Real >::NumCols(), MatrixBase< Real >::NumRows(), ParseOptions::PrintUsage(), ParseOptions::Read(), kaldi::ReadKaldiObject(), ParseOptions::Register(), kaldi::ShiftFeatureMatrix(), and kaldi::WriteKaldiObject().

                                 {
  try {
    using namespace kaldi;

    const char *usage =
        "Copy features, and possibly shift them while maintaining the "
        "num-frames.\n"
        "Usage: shift-feats [options] <feature-rspecifier> "
        "<feature-wspecifier>\n"
        "or:  shift-feats [options] <feats-rxfilename> <feats-wxfilename>\n"
        "e.g.: shift-feats --shift=-1 foo.scp bar.ark\n"
        "or: shift-feats --shift=1 foo.mat bar.mat\n"
        "See also: copy-feats, copy-matrix, select-feats, subset-feats,\n"
        "subsample-feats, splice-feats, paste-feats, concat-feats, "
        "extract-feature-segments\n";

    ParseOptions po(usage);
    bool binary = true;
    int32 shift = 0;
    po.Register("shift", &shift, "Number of frames by which to shift the "
                                 "features.");
    po.Register("binary", &binary, "Binary-mode output (not relevant if "
                "writing to archive)");

    po.Read(argc, argv);

    if (po.NumArgs() != 2) {
      po.PrintUsage();
      exit(1);
    }

    int32 num_done = 0, num_err = 0;

    if (ClassifyRspecifier(po.GetArg(1), NULL, NULL) != kNoRspecifier) {
      SequentialBaseFloatMatrixReader feat_reader(po.GetArg(1));
      BaseFloatMatrixWriter feat_writer(po.GetArg(2));


      for (; !feat_reader.Done(); feat_reader.Next()) {
        const std::string &key = feat_reader.Key();
        const Matrix<BaseFloat> &src = feat_reader.Value();
        if (src.NumRows() == 0) {
          KALDI_WARN << "Empty matrix for key " << key;
          num_err++;
          continue;
        }
        Matrix<BaseFloat> rearranged(src.NumRows(), src.NumCols());
        ShiftFeatureMatrix(src, shift, &rearranged);
        feat_writer.Write(key, rearranged);
        num_done++;
      }

      KALDI_LOG << "Shifted " << num_done << " features by "
                << shift << " frames; " << num_err << " with errors.";
      return (num_done > 0 ? 0 : 1);
    } else {
      std::string feat_rxfilename = po.GetArg(1),
                  feat_wxfilename = po.GetArg(2);
      Matrix<BaseFloat> src;
      ReadKaldiObject(feat_rxfilename, &src);
      if (src.NumRows() == 0)
        KALDI_ERR << "Empty input matrix";
      Matrix<BaseFloat> rearranged(src.NumRows(), src.NumCols());
      ShiftFeatureMatrix(src, shift, &rearranged);
      WriteKaldiObject(rearranged, feat_wxfilename, binary);
      // we do not print any log messages here
    }
  } catch(const std::exception &e) {
    std::cerr << e.what();
    return -1;
  }
}