fstrmsymbols.cc File Reference

#include "base/kaldi-common.h"
#include "util/common-utils.h"
#include "fst/fstlib.h"
#include "fstext/determinize-star.h"
#include "fstext/fstext-utils.h"
#include "fstext/kaldi-fst-io.h"

Include dependency graph for fstrmsymbols.cc:

Go to the source code of this file.

Namespaces
	fst
	For an extended explanation of the framework of which grammar-fsts are a part, please see Support for grammars and graphs with on-the-fly parts. (i.e.

Functions
template<class Arc , class I >
void	RemoveArcsWithSomeInputSymbols (const std::vector< I > &symbols_in, VectorFst< Arc > *fst)
template<class Arc , class I >
void	PenalizeArcsWithSomeInputSymbols (const std::vector< I > &symbols_in, float penalty, VectorFst< Arc > *fst)
int	main (int argc, char *argv[])

Function Documentation

main()

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

Definition at line 85 of file fstrmsymbols.cc.

References fst::CastOrConvertToVectorFst(), ParseOptions::GetArg(), ParseOptions::GetOptArg(), KALDI_ERR, ParseOptions::NumArgs(), fst::PenalizeArcsWithSomeInputSymbols(), ParseOptions::PrintUsage(), ParseOptions::Read(), fst::ReadFstKaldiGeneric(), kaldi::ReadIntegerVectorSimple(), ParseOptions::Register(), fst::RemoveArcsWithSomeInputSymbols(), fst::RemoveSomeInputSymbols(), and fst::WriteFstKaldi().

                                 {
  try {
    using namespace kaldi;
    using namespace fst;
    using kaldi::int32;

    bool apply_to_output = false;
    bool remove_arcs = false;
    float penalty = -std::numeric_limits<BaseFloat>::infinity();

    const char *usage =
        "With no options, replaces a subset of symbols with epsilon, wherever\n"
        "they appear on the input side of an FST."
        "With --remove-arcs=true, will remove arcs that contain these symbols\n"
        "on the input\n"
        "With --penalty=<float>, will add the specified penalty to the\n"
        "cost of any arc that has one of the given symbols on its input side\n"
        "In all cases, the option --apply-to-output=true (or for\n"
        "back-compatibility, --remove-from-output=true) makes this apply\n"
        "to the output side.\n"
        "\n"
        "Usage:  fstrmsymbols [options] <in-disambig-list>  [<in.fst> [<out.fst>]]\n"
        "E.g:  fstrmsymbols in.list  < in.fst > out.fst\n"
        "<in-disambig-list> is an rxfilename specifying a file containing list of integers\n"
        "representing symbols, in text form, one per line.\n";

    ParseOptions po(usage);
    po.Register("remove-from-output", &apply_to_output, "If true, this applies to symbols "
                "on the output, not the input, side.  (For back compatibility; use "
                "--apply-to-output insead)");
    po.Register("apply-to-output", &apply_to_output, "If true, this applies to symbols "
                "on the output, not the input, side.");
    po.Register("remove-arcs", &remove_arcs, "If true, instead of converting the symbol "
                "to <eps>, remove the arcs.");
    po.Register("penalty", &penalty, "If specified, instead of converting "
                "the symbol to <eps>, penalize the arc it is on by adding this "
                "value to its cost.");


    po.Read(argc, argv);

    if (remove_arcs &&
        penalty != -std::numeric_limits<BaseFloat>::infinity())
      KALDI_ERR << "--remove-arc and --penalty options are mutually exclusive";

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

    std::string disambig_rxfilename = po.GetArg(1),
        fst_rxfilename = po.GetOptArg(2),
        fst_wxfilename = po.GetOptArg(3);

    VectorFst<StdArc> *fst = CastOrConvertToVectorFst(
        ReadFstKaldiGeneric(fst_rxfilename));

    std::vector<int32> disambig_in;
    if (!ReadIntegerVectorSimple(disambig_rxfilename, &disambig_in))
      KALDI_ERR << "fstrmsymbols: Could not read disambiguation symbols from "
                << (disambig_rxfilename == "" ? "standard input" : disambig_rxfilename);

    if (apply_to_output) Invert(fst);
    if (remove_arcs) {
      RemoveArcsWithSomeInputSymbols(disambig_in, fst);
    } else if (penalty != -std::numeric_limits<BaseFloat>::infinity()) {
      PenalizeArcsWithSomeInputSymbols(disambig_in, penalty, fst);
    } else {
      RemoveSomeInputSymbols(disambig_in, fst);
    }
    if (apply_to_output) Invert(fst);

    WriteFstKaldi(*fst, fst_wxfilename);

    delete fst;
    return 0;
  } catch(const std::exception &e) {
    std::cerr << e.what();
    return -1;
  }
}