All Classes Namespaces Files Functions Variables Typedefs Enumerations Enumerator Friends Macros Modules Pages
compile-train-graphs.cc File Reference
Include dependency graph for compile-train-graphs.cc:

Go to the source code of this file.

Functions

int main (int argc, char *argv[])
 

Function Documentation

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

Definition at line 29 of file compile-train-graphs.cc.

References TrainingGraphCompiler::CompileGraphFromText(), TrainingGraphCompiler::CompileGraphsFromText(), kaldi::DeletePointers(), SequentialTableReader< Holder >::Done(), ParseOptions::GetArg(), rnnlm::i, KALDI_ASSERT, KALDI_ERR, KALDI_LOG, KALDI_WARN, SequentialTableReader< Holder >::Key(), SequentialTableReader< Holder >::Next(), ParseOptions::NumArgs(), ParseOptions::PrintUsage(), ParseOptions::Read(), fst::ReadFstKaldi(), kaldi::ReadIntegerVectorSimple(), kaldi::ReadKaldiObject(), TrainingGraphCompilerOptions::Register(), ParseOptions::Register(), TrainingGraphCompilerOptions::self_loop_scale, TrainingGraphCompilerOptions::transition_scale, SequentialTableReader< Holder >::Value(), and TableWriter< Holder >::Write().

29  {
30  try {
31  using namespace kaldi;
32  typedef kaldi::int32 int32;
33  using fst::SymbolTable;
34  using fst::VectorFst;
35  using fst::StdArc;
36 
37  const char *usage =
38  "Creates training graphs (without transition-probabilities, by default)\n"
39  "\n"
40  "Usage: compile-train-graphs [options] <tree-in> <model-in> "
41  "<lexicon-fst-in> <transcriptions-rspecifier> <graphs-wspecifier>\n"
42  "e.g.: \n"
43  " compile-train-graphs tree 1.mdl lex.fst "
44  "'ark:sym2int.pl -f 2- words.txt text|' ark:graphs.fsts\n";
45  ParseOptions po(usage);
46 
48  int32 batch_size = 250;
49  gopts.transition_scale = 0.0; // Change the default to 0.0 since we will generally add the
50  // transition probs in the alignment phase (since they change eacm time)
51  gopts.self_loop_scale = 0.0; // Ditto for self-loop probs.
52  std::string disambig_rxfilename;
53  gopts.Register(&po);
54 
55  po.Register("batch-size", &batch_size,
56  "Number of FSTs to compile at a time (more -> faster but uses "
57  "more memory. E.g. 500");
58  po.Register("read-disambig-syms", &disambig_rxfilename, "File containing "
59  "list of disambiguation symbols in phone symbol table");
60 
61  po.Read(argc, argv);
62 
63  if (po.NumArgs() != 5) {
64  po.PrintUsage();
65  exit(1);
66  }
67 
68  std::string tree_rxfilename = po.GetArg(1);
69  std::string model_rxfilename = po.GetArg(2);
70  std::string lex_rxfilename = po.GetArg(3);
71  std::string transcript_rspecifier = po.GetArg(4);
72  std::string fsts_wspecifier = po.GetArg(5);
73 
74  ContextDependency ctx_dep; // the tree.
75  ReadKaldiObject(tree_rxfilename, &ctx_dep);
76 
77  TransitionModel trans_model;
78  ReadKaldiObject(model_rxfilename, &trans_model);
79 
80  // need VectorFst because we will change it by adding subseq symbol.
81  VectorFst<StdArc> *lex_fst = fst::ReadFstKaldi(lex_rxfilename);
82 
83  std::vector<int32> disambig_syms;
84  if (disambig_rxfilename != "")
85  if (!ReadIntegerVectorSimple(disambig_rxfilename, &disambig_syms))
86  KALDI_ERR << "fstcomposecontext: Could not read disambiguation symbols from "
87  << disambig_rxfilename;
88 
89  TrainingGraphCompiler gc(trans_model, ctx_dep, lex_fst, disambig_syms, gopts);
90 
91  lex_fst = NULL; // we gave ownership to gc.
92 
93  SequentialInt32VectorReader transcript_reader(transcript_rspecifier);
94  TableWriter<fst::VectorFstHolder> fst_writer(fsts_wspecifier);
95 
96  int num_succeed = 0, num_fail = 0;
97 
98  if (batch_size == 1) { // We treat batch_size of 1 as a special case in order
99  // to test more parts of the code.
100  for (; !transcript_reader.Done(); transcript_reader.Next()) {
101  std::string key = transcript_reader.Key();
102  const std::vector<int32> &transcript = transcript_reader.Value();
103  VectorFst<StdArc> decode_fst;
104 
105  if (!gc.CompileGraphFromText(transcript, &decode_fst)) {
106  decode_fst.DeleteStates(); // Just make it empty.
107  }
108  if (decode_fst.Start() != fst::kNoStateId) {
109  num_succeed++;
110  fst_writer.Write(key, decode_fst);
111  } else {
112  KALDI_WARN << "Empty decoding graph for utterance "
113  << key;
114  num_fail++;
115  }
116  }
117  } else {
118  std::vector<std::string> keys;
119  std::vector<std::vector<int32> > transcripts;
120  while (!transcript_reader.Done()) {
121  keys.clear();
122  transcripts.clear();
123  for (; !transcript_reader.Done() &&
124  static_cast<int32>(transcripts.size()) < batch_size;
125  transcript_reader.Next()) {
126  keys.push_back(transcript_reader.Key());
127  transcripts.push_back(transcript_reader.Value());
128  }
129  std::vector<fst::VectorFst<fst::StdArc>* > fsts;
130  if (!gc.CompileGraphsFromText(transcripts, &fsts)) {
131  KALDI_ERR << "Not expecting CompileGraphs to fail.";
132  }
133  KALDI_ASSERT(fsts.size() == keys.size());
134  for (size_t i = 0; i < fsts.size(); i++) {
135  if (fsts[i]->Start() != fst::kNoStateId) {
136  num_succeed++;
137  fst_writer.Write(keys[i], *(fsts[i]));
138  } else {
139  KALDI_WARN << "Empty decoding graph for utterance "
140  << keys[i];
141  num_fail++;
142  }
143  }
144  DeletePointers(&fsts);
145  }
146  }
147  KALDI_LOG << "compile-train-graphs: succeeded for " << num_succeed
148  << " graphs, failed for " << num_fail;
149  return (num_succeed != 0 ? 0 : 1);
150  } catch(const std::exception &e) {
151  std::cerr << e.what();
152  return -1;
153  }
154 }
Relabels neural network egs with the read pdf-id alignments.
Definition: chain.dox:20
fst::StdArc StdArc
A templated class for writing objects to an archive or script file; see The Table concept...
Definition: kaldi-table.h:366
void ReadKaldiObject(const std::string &filename, Matrix< float > *m)
Definition: kaldi-io.cc:818
The class ParseOptions is for parsing command-line options; see Parsing command-line options for more...
Definition: parse-options.h:36
A templated class for reading objects sequentially from an archive or script file; see The Table conc...
Definition: kaldi-table.h:287
#define KALDI_ERR
Definition: kaldi-error.h:127
#define KALDI_WARN
Definition: kaldi-error.h:130
#define KALDI_ASSERT(cond)
Definition: kaldi-error.h:169
void ReadFstKaldi(std::istream &is, bool binary, VectorFst< Arc > *fst)
#define KALDI_LOG
Definition: kaldi-error.h:133
void DeletePointers(std::vector< A * > *v)
Deletes any non-NULL pointers in the vector v, and sets the corresponding entries of v to NULL...
Definition: stl-utils.h:186
bool ReadIntegerVectorSimple(std::string rxfilename, std::vector< int32 > *list)
ReadFromList attempts to read this list of integers, one per line, from the given file...