compile-questions.cc File Reference

#include "base/kaldi-common.h"
#include "util/common-utils.h"
#include "hmm/hmm-topology.h"
#include "tree/build-tree-questions.h"

Include dependency graph for compile-questions.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
int32	ProcessTopo (const HmmTopology &topo, const std::vector< std::vector< int32 > > &questions)
int	main (int argc, char *argv[])

Function Documentation

main()

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

Definition at line 57 of file compile-questions.cc.

References ParseOptions::GetArg(), rnnlm::i, QuestionsForKey::initial_questions, kaldi::IsSortedAndUniq(), rnnlm::j, KALDI_ERR, KALDI_LOG, KALDI_WARN, kaldi::kPdfClass, rnnlm::n, ParseOptions::NumArgs(), kaldi::PrintableRxfilename(), ParseOptions::PrintUsage(), kaldi::ProcessTopo(), ParseOptions::Read(), kaldi::ReadIntegerVectorVectorSimple(), kaldi::ReadKaldiObject(), ParseOptions::Register(), Questions::SetQuestionsOf(), kaldi::SortAndUniq(), and kaldi::WriteKaldiObject().

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

    const char *usage =
        "Compile questions\n"
        "Usage:  compile-questions [options] <topo> <questions-text-file> <questions-out>\n"
        "e.g.: \n"
        " compile-questions questions.txt questions.qst\n";
    bool binary = true;
    int32 P = 1, N = 3;
    int32 num_iters_refine = 0;


    ParseOptions po(usage);
    po.Register("binary", &binary,
                "Write output in binary mode");
    po.Register("context-width", &N,
                "Context window size [must match acc-tree-stats].");
    po.Register("central-position", &P,
                "Central position in phone context window [must match acc-tree-stats]");
    po.Register("num-iters-refine", &num_iters_refine,
                "Number of iters of refining questions at each node.  >0 --> questions "
                "not refined");

    po.Read(argc, argv);

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

    std::string
        topo_filename = po.GetArg(1),
        questions_rxfilename = po.GetArg(2),
        questions_out_filename = po.GetArg(3);

    HmmTopology topo;  // just needed for checking, and to get the
    // largest number of pdf-classes for any phone.
    ReadKaldiObject(topo_filename, &topo);

    std::vector<std::vector<int32> > questions;  // sets of phones.
    if (!ReadIntegerVectorVectorSimple(questions_rxfilename, &questions))
      KALDI_ERR << "Could not read questions from "
                 << PrintableRxfilename(questions_rxfilename);
    for (size_t i = 0; i < questions.size(); i++) {
      std::sort(questions[i].begin(), questions[i].end());
      if (!IsSortedAndUniq(questions[i]))
        KALDI_ERR << "Questions contain duplicate phones";
    }
    size_t nq = static_cast<int32>(questions.size());
    SortAndUniq(&questions);
    if (questions.size() != nq)
      KALDI_WARN << (nq-questions.size())
                 << " duplicate questions present in " << questions_rxfilename;

    // ProcessTopo checks that all phones in the topo are
    // represented in at least one questions (else warns), and
    // returns the max # pdf classes in any given phone (normally
    // 3).
    int32 max_num_pdfclasses = ProcessTopo(topo, questions);

    Questions qo;

    QuestionsForKey phone_opts(num_iters_refine);
    // the questions-options corresponding to keys 0, 1, .. N-1 which
    // represent the phonetic context positions (including the central phone).
    for (int32 n = 0; n < N; n++) {
      KALDI_LOG << "Setting questions for phonetic-context position "<< n;
      phone_opts.initial_questions = questions;
      qo.SetQuestionsOf(n, phone_opts);
    }

    QuestionsForKey pdfclass_opts(num_iters_refine);
    std::vector<std::vector<int32> > pdfclass_questions(max_num_pdfclasses-1);
    for (int32 i = 0; i < max_num_pdfclasses - 1; i++)
      for (int32 j = 0; j <= i; j++)
        pdfclass_questions[i].push_back(j);
    // E.g. if max_num_pdfclasses == 3,  pdfclass_questions is now [ [0], [0, 1] ].
    pdfclass_opts.initial_questions = pdfclass_questions;
    KALDI_LOG << "Setting questions for hmm-position [hmm-position ranges from 0 to "<< (max_num_pdfclasses-1) <<"]";
    qo.SetQuestionsOf(kPdfClass, pdfclass_opts);

    WriteKaldiObject(qo, questions_out_filename, binary);
    KALDI_LOG << "Wrote questions to "<<questions_out_filename;
  } catch(const std::exception &e) {
    std::cerr << e.what();
    return -1;
  }
}