paste-feats.cc File Reference

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

Include dependency graph for paste-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
bool	AppendFeats (const std::vector< Matrix< BaseFloat > > &in, const std::string &utt, int32 tolerance, Matrix< BaseFloat > *out)
int	main (int argc, char *argv[])

Function Documentation

main()

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

Definition at line 70 of file paste-feats.cc.

References kaldi::AppendFeats(), kaldi::ClassifyRspecifier(), SequentialTableReader< Holder >::Done(), ParseOptions::GetArg(), rnnlm::i, KALDI_LOG, KALDI_VLOG, KALDI_WARN, SequentialTableReader< Holder >::Key(), kaldi::kNoRspecifier, SequentialTableReader< Holder >::Next(), ParseOptions::NumArgs(), ParseOptions::PrintUsage(), ParseOptions::Read(), kaldi::ReadKaldiObject(), ParseOptions::Register(), SequentialTableReader< Holder >::Value(), TableWriter< Holder >::Write(), and kaldi::WriteKaldiObject().

                                 {
  try {
    using namespace kaldi;
    using namespace std;

    const char *usage =
        "Paste feature files (assuming they have about the same durations,\n"
        "see --length-tolerance), appending the features on each frame;\n"
        "think of the unix command 'paste'.\n"
        "Usage: paste-feats <in-rspecifier1> <in-rspecifier2> [<in-rspecifier3> ...] <out-wspecifier>\n"
        " or: paste-feats <in-rxfilename1> <in-rxfilename2> [<in-rxfilename3> ...] <out-wxfilename>\n"
        " e.g. paste-feats ark:feats1.ark \"ark:select-feats 0-3 ark:feats2.ark ark:- |\" ark:feats-out.ark\n"
        "  or: paste-feats foo.mat bar.mat baz.mat\n"
        "See also: copy-feats, copy-matrix, append-vector-to-feats, concat-feats\n";

    ParseOptions po(usage);

    int32 length_tolerance = 0;
    bool binary = true;
    po.Register("length-tolerance", &length_tolerance,
                "If length is different, trim as shortest up to a frame "
                " difference of length-tolerance, otherwise exclude segment.");
    po.Register("binary", &binary, "If true, output files in binary "
                "(only relevant for single-file operation, i.e. no tables)");

    po.Read(argc, argv);

    if (po.NumArgs() < 3) {
      po.PrintUsage();
      exit(1);
    }

    if (ClassifyRspecifier(po.GetArg(1), NULL, NULL)
        != kNoRspecifier) {
      // We're operating on tables, e.g. archives.

      // Last argument is output
      string wspecifier = po.GetArg(po.NumArgs());
      BaseFloatMatrixWriter feat_writer(wspecifier);

      // First input is sequential
      string rspecifier1 = po.GetArg(1);
      SequentialBaseFloatMatrixReader input1(rspecifier1);

      // Assemble vector of other input readers (with random-access)
      vector<RandomAccessBaseFloatMatrixReader *> input;
      for (int32 i = 2; i < po.NumArgs(); i++) {
        string rspecifier = po.GetArg(i);
        RandomAccessBaseFloatMatrixReader *rd = new RandomAccessBaseFloatMatrixReader(rspecifier);
        input.push_back(rd);
      }

      int32 num_done = 0, num_err = 0;

      // Main loop
      for (; !input1.Done(); input1.Next()) {
        string utt = input1.Key();
        KALDI_VLOG(2) << "Merging features for utterance " << utt;

        // Collect features from streams to vector 'feats'
        vector<Matrix<BaseFloat> > feats(po.NumArgs() - 1);
        feats[0] = input1.Value();
        int32 i;
        for (i = 0; i < static_cast<int32>(input.size()); i++) {
          if (input[i]->HasKey(utt)) {
            feats[i + 1] = input[i]->Value(utt);
          } else {
            KALDI_WARN << "Missing utt " << utt << " from input "
                       << po.GetArg(i+2);
            num_err++;
            break;
          }
        }
        if (i != static_cast<int32>(input.size()))
          continue;
        Matrix<BaseFloat> output;
        if (!AppendFeats(feats, utt, length_tolerance, &output)) {
          num_err++;
          continue; // it will have printed a warning.
        }
        feat_writer.Write(utt, output);
        num_done++;
      }

      for (int32 i=0; i < input.size(); i++)
        delete input[i];
      input.clear();

      KALDI_LOG << "Done " << num_done << " utts, errors on "
                << num_err;

      return (num_done == 0 ? -1 : 0);
    } else {
      // We're operating on rxfilenames|wxfilenames, most likely files.
      std::vector<Matrix<BaseFloat> > feats(po.NumArgs() - 1);
      for (int32 i = 1; i < po.NumArgs(); i++)
        ReadKaldiObject(po.GetArg(i), &(feats[i-1]));
      Matrix<BaseFloat> output;
      if (!AppendFeats(feats, "", length_tolerance, &output))
        return 1; // it will have printed a warning.
      std::string output_wxfilename = po.GetArg(po.NumArgs());
      WriteKaldiObject(output, output_wxfilename, binary);
      KALDI_LOG << "Wrote appended features to " << output_wxfilename;
      return 0;
    }
  } catch(const std::exception &e) {
    std::cerr << e.what();
    return -1;
  }
}