nnet3-discriminative-get-egs.cc File Reference

#include <sstream>
#include "base/kaldi-common.h"
#include "util/common-utils.h"
#include "hmm/transition-model.h"
#include "hmm/posterior.h"
#include "nnet3/nnet-discriminative-example.h"
#include "nnet3/discriminative-supervision.h"
#include "nnet3/nnet-example-utils.h"
#include "chain/chain-supervision.h"

Include dependency graph for nnet3-discriminative-get-egs.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:
	kaldi::nnet3

Functions
static bool	ProcessFile (const discriminative::SplitDiscriminativeSupervisionOptions &config, const TransitionModel &tmodel, const MatrixBase< BaseFloat > &feats, const MatrixBase< BaseFloat > *ivector_feats, int32 ivector_period, const discriminative::DiscriminativeSupervision &supervision, const std::string &utt_id, bool compress, UtteranceSplitter *utt_splitter, NnetDiscriminativeExampleWriter *example_writer)
int	main (int argc, char *argv[])

Function Documentation

main()

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

Definition at line 155 of file nnet3-discriminative-get-egs.cc.

References ExampleGenerationConfig::ComputeDerived(), SequentialTableReader< Holder >::Done(), UtteranceSplitter::ExitStatus(), ParseOptions::GetArg(), RandomAccessTableReader< Holder >::HasKey(), DiscriminativeSupervision::Initialize(), KALDI_WARN, SequentialTableReader< Holder >::Key(), SequentialTableReader< Holder >::Next(), ParseOptions::NumArgs(), MatrixBase< Real >::NumRows(), ParseOptions::PrintUsage(), kaldi::nnet3::ProcessFile(), ParseOptions::Read(), TransitionModel::Read(), SplitDiscriminativeSupervisionOptions::Register(), ParseOptions::Register(), ExampleGenerationConfig::Register(), Input::Stream(), RandomAccessTableReader< Holder >::Value(), and SequentialTableReader< Holder >::Value().

                                 {
  try {
    using namespace kaldi;
    using namespace kaldi::nnet3;
    typedef kaldi::int32 int32;
    typedef kaldi::int64 int64;

    const char *usage =
        "Get frame-by-frame examples of data for nnet3+sequence neural network\n"
        "training.  This involves breaking up utterances into pieces of sizes\n"
        "determined by the --num-frames option.\n"
        "\n"
        "Usage:  nnet3-discriminative-get-egs [options] <model> <features-rspecifier> "
        "<denominator-lattice-rspecifier> <numerator-alignment-rspecifier> <egs-wspecifier>\n"
        "\n"
        "An example [where $feats expands to the actual features]:\n"
        "  nnet3-discriminative-get-egs --left-context=25 --right-context=9 --num-frames=150,100,90 \\\n"
        "  \"$feats\" \"ark,s,cs:gunzip -c lat.1.gz\" scp:ali.scp ark:degs.1.ark\n";

    bool compress = true;
    int32 length_tolerance = 100, online_ivector_period = 1;

    std::string online_ivector_rspecifier;

    ExampleGenerationConfig eg_config;  // controls num-frames,
                                        // left/right-context, etc.
    discriminative::SplitDiscriminativeSupervisionOptions splitter_config;

    ParseOptions po(usage);

    eg_config.Register(&po);
    po.Register("compress", &compress, "If true, write egs in "
                "compressed format (recommended)");
    po.Register("ivectors", &online_ivector_rspecifier, "Alias for --online-ivectors "
                "option, for back compatibility");
    po.Register("online-ivectors", &online_ivector_rspecifier, "Rspecifier of ivector "
                "features, as a matrix.");
    po.Register("online-ivector-period", &online_ivector_period, "Number of frames "
                "between iVectors in matrices supplied to the --online-ivectors "
                "option");
    po.Register("length-tolerance", &length_tolerance, "Tolerance for "
                "difference in num-frames between feat and ivector matrices");

    splitter_config.Register(&po);

    po.Read(argc, argv);

    if (po.NumArgs() != 5) {
      po.PrintUsage();
      exit(1);
    }

    eg_config.ComputeDerived();
    UtteranceSplitter utt_splitter(eg_config);

    std::string model_wxfilename = po.GetArg(1),
        feature_rspecifier = po.GetArg(2),
        den_lat_rspecifier = po.GetArg(3),
        num_ali_rspecifier = po.GetArg(4),
        examples_wspecifier = po.GetArg(5);


    TransitionModel tmodel;
    {
      bool binary;
      Input ki(model_wxfilename, &binary);
      tmodel.Read(ki.Stream(), binary);
    }

    SequentialBaseFloatMatrixReader feat_reader(feature_rspecifier);
    RandomAccessLatticeReader den_lat_reader(den_lat_rspecifier);
    RandomAccessInt32VectorReader ali_reader(num_ali_rspecifier);
    NnetDiscriminativeExampleWriter example_writer(examples_wspecifier);
    RandomAccessBaseFloatMatrixReader online_ivector_reader(
        online_ivector_rspecifier);

    int32 num_err = 0;

    for (; !feat_reader.Done(); feat_reader.Next()) {
      std::string key = feat_reader.Key();
      const Matrix<BaseFloat> &feats = feat_reader.Value();
      if (!den_lat_reader.HasKey(key)) {
        KALDI_WARN << "No denominator lattice for key " << key;
        num_err++;
      } else if (!ali_reader.HasKey(key)) {
        KALDI_WARN << "No numerator alignment for key " << key;
        num_err++;
      } else {
        discriminative::DiscriminativeSupervision supervision;
        if (!supervision.Initialize(ali_reader.Value(key),
                                    den_lat_reader.Value(key),
                                    1.0)) {
          KALDI_WARN << "Failed to convert lattice to supervision "
                     << "for utterance " << key;
          num_err++;
          continue;
        }

        const Matrix<BaseFloat> *online_ivector_feats = NULL;
        if (!online_ivector_rspecifier.empty()) {
          if (!online_ivector_reader.HasKey(key)) {
            KALDI_WARN << "No iVectors for utterance " << key;
            num_err++;
            continue;
          } else {
            // this address will be valid until we call HasKey() or Value()
            // again.
            online_ivector_feats = &(online_ivector_reader.Value(key));
          }
        }
        if (online_ivector_feats != NULL &&
            (abs(feats.NumRows() - (online_ivector_feats->NumRows() *
                                    online_ivector_period)) > length_tolerance
             || online_ivector_feats->NumRows() == 0)) {
          KALDI_WARN << "Length difference between feats " << feats.NumRows()
                     << " and iVectors " << online_ivector_feats->NumRows()
                     << "exceeds tolerance " << length_tolerance;
          num_err++;
          continue;
        }
        if (!ProcessFile(splitter_config, tmodel,
                         feats, online_ivector_feats, online_ivector_period,
                         supervision, key, compress,
                         &utt_splitter, &example_writer))
          num_err++;
      }
    }
    if (num_err > 0)
      KALDI_WARN << num_err << " utterances had errors and could "
          "not be processed.";
    // utt_splitter prints diagnostics.
    return utt_splitter.ExitStatus();
  } catch(const std::exception &e) {
    std::cerr << e.what() << '\n';
    return -1;
  }
}