lattice-simple-decoder.cc

Go to the documentation of this file.

// decoder/lattice-simple-decoder.cc

// Copyright 2009-2012  Microsoft Corporation
//           2013-2014  Johns Hopkins University (Author: Daniel Povey)
//                2014  Guoguo Chen

// See ../../COPYING for clarification regarding multiple authors
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
//  http://www.apache.org/licenses/LICENSE-2.0
//
// THIS CODE IS PROVIDED *AS IS* BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
// KIND, EITHER EXPRESS OR IMPLIED, INCLUDING WITHOUT LIMITATION ANY IMPLIED
// WARRANTIES OR CONDITIONS OF TITLE, FITNESS FOR A PARTICULAR PURPOSE,
// MERCHANTABLITY OR NON-INFRINGEMENT.
// See the Apache 2 License for the specific language governing permissions and
// limitations under the License.

#include "lattice-simple-decoder.h"

namespace kaldi {


void LatticeSimpleDecoder::InitDecoding() {
  // clean up from last time:
  cur_toks_.clear();
  prev_toks_.clear();
  ClearActiveTokens();
  warned_ = false;
  decoding_finalized_ = false;
  final_costs_.clear();
  num_toks_ = 0;
  StateId start_state = fst_.Start();
  KALDI_ASSERT(start_state != fst::kNoStateId);
  active_toks_.resize(1);
  Token *start_tok = new Token(0.0, 0.0, NULL, NULL);
  active_toks_[0].toks = start_tok;
  cur_toks_[start_state] = start_tok;
  num_toks_++;
  ProcessNonemitting();
}

bool LatticeSimpleDecoder::Decode(DecodableInterface *decodable) {
  InitDecoding();
  
  while (!decodable->IsLastFrame(NumFramesDecoded() - 1)) {  
    if (NumFramesDecoded() % config_.prune_interval == 0)
      PruneActiveTokens(config_.lattice_beam * config_.prune_scale);
    ProcessEmitting(decodable);
    // Important to call PruneCurrentTokens before ProcessNonemitting, or we
    // would get dangling forward pointers.  Anyway, ProcessNonemitting uses the
    // beam.
    PruneCurrentTokens(config_.beam, &cur_toks_);
    ProcessNonemitting();
  }
  FinalizeDecoding();
  
  // Returns true if we have any kind of traceback available (not necessarily
  // to the end state; query ReachedFinal() for that).
  return !final_costs_.empty();
}

// Outputs an FST corresponding to the single best path
// through the lattice.
bool LatticeSimpleDecoder::GetBestPath(Lattice *ofst,
                                       bool use_final_probs) const {
  fst::VectorFst<LatticeArc> fst;
  GetRawLattice(&fst, use_final_probs);
  ShortestPath(fst, ofst);
  return (ofst->NumStates() > 0);
}

// Outputs an FST corresponding to the raw, state-level
// tracebacks.
bool LatticeSimpleDecoder::GetRawLattice(Lattice *ofst,
                                         bool use_final_probs) const {
  typedef LatticeArc Arc;
  typedef Arc::StateId StateId;
  typedef Arc::Weight Weight;
  typedef Arc::Label Label;

  // Note: you can't use the old interface (Decode()) if you want to
  // get the lattice with use_final_probs = false.  You'd have to do
  // InitDecoding() and then AdvanceDecoding().
  if (decoding_finalized_ && !use_final_probs)
    KALDI_ERR << "You cannot call FinalizeDecoding() and then call "
              << "GetRawLattice() with use_final_probs == false";
  
  unordered_map<Token*, BaseFloat> final_costs_local;
  
  const unordered_map<Token*, BaseFloat> &final_costs =
      (decoding_finalized_ ? final_costs_ : final_costs_local);

  if (!decoding_finalized_ && use_final_probs)
    ComputeFinalCosts(&final_costs_local, NULL, NULL);

  ofst->DeleteStates();
  int32 num_frames = NumFramesDecoded();
  KALDI_ASSERT(num_frames > 0);
  const int32 bucket_count = num_toks_/2 + 3;  
  unordered_map<Token*, StateId> tok_map(bucket_count);
  // First create all states.
  for (int32 f = 0; f <= num_frames; f++) {
    if (active_toks_[f].toks == NULL) {
      KALDI_WARN << "GetRawLattice: no tokens active on frame " << f
                 << ": not producing lattice.\n";
      return false;
    }
    for (Token *tok = active_toks_[f].toks; tok != NULL; tok = tok->next)
      tok_map[tok] = ofst->AddState();
    // The next statement sets the start state of the output FST.
    // Because we always add new states to the head of the list
    // active_toks_[f].toks, and the start state was the first one
    // added, it will be the last one added to ofst.
    if (f == 0 && ofst->NumStates() > 0)
      ofst->SetStart(ofst->NumStates()-1);
  }
  StateId cur_state = 0; // we rely on the fact that we numbered these
  // consecutively (AddState() returns the numbers in order..)
  for (int32 f = 0; f <= num_frames; f++) {
    for (Token *tok = active_toks_[f].toks; tok != NULL; tok = tok->next,
             cur_state++) {
      for (ForwardLink *l = tok->links;
           l != NULL;
           l = l->next) {
        unordered_map<Token*, StateId>::const_iterator iter =
            tok_map.find(l->next_tok);
        StateId nextstate = iter->second;
        KALDI_ASSERT(iter != tok_map.end());
        Arc arc(l->ilabel, l->olabel,
                Weight(l->graph_cost, l->acoustic_cost),
                nextstate);
        ofst->AddArc(cur_state, arc);
      }
      if (f == num_frames) {
        if (use_final_probs && !final_costs.empty()) {
          unordered_map<Token*, BaseFloat>::const_iterator iter =
              final_costs.find(tok);
          if (iter != final_costs.end())
            ofst->SetFinal(cur_state, LatticeWeight(iter->second, 0));
        } else {
          ofst->SetFinal(cur_state, LatticeWeight::One());
        }
      }
    }
  }
  KALDI_ASSERT(cur_state == ofst->NumStates());
  return (cur_state != 0);
}

// This function is now deprecated, since now we do determinization from outside
// the LatticeSimpleDecoder class.
// Outputs an FST corresponding to the lattice-determinized
// lattice (one path per word sequence).
bool LatticeSimpleDecoder::GetLattice(
    CompactLattice *ofst,
    bool use_final_probs) const {
  Lattice raw_fst;
  GetRawLattice(&raw_fst, use_final_probs);
  Invert(&raw_fst); // make it so word labels are on the input.
  if (!TopSort(&raw_fst)) // topological sort makes lattice-determinization more efficient
    KALDI_WARN << "Topological sorting of state-level lattice failed "
        "(probably your lexicon has empty words or your LM has epsilon cycles; this "
        " is a bad idea.)";
  // (in phase where we get backward-costs).
  fst::ILabelCompare<LatticeArc> ilabel_comp;
  ArcSort(&raw_fst, ilabel_comp); // sort on ilabel; makes
  // lattice-determinization more efficient.
    
  fst::DeterminizeLatticePrunedOptions lat_opts;
  lat_opts.max_mem = config_.det_opts.max_mem;
    
  DeterminizeLatticePruned(raw_fst, config_.lattice_beam, ofst, lat_opts);
  raw_fst.DeleteStates(); // Free memory-- raw_fst no longer needed.
  Connect(ofst); // Remove unreachable states... there might be
  // a small number of these, in some cases.
  // Note: if something went wrong and the raw lattice was empty,
  // we should still get to this point in the code without warnings or failures.
  return (ofst->NumStates() != 0);
}



// FindOrAddToken either locates a token in cur_toks_, or if necessary inserts a new,
// empty token (i.e. with no forward links) for the current frame.  [note: it's
// inserted if necessary into cur_toks_ and also into the singly linked list
// of tokens active on this frame (whose head is at active_toks_[frame]).
//
// Returns the Token pointer.  Sets "changed" (if non-NULL) to true
// if the token was newly created or the cost changed.
inline LatticeSimpleDecoder::Token *LatticeSimpleDecoder::FindOrAddToken(
    StateId state, int32 frame, BaseFloat tot_cost,
    bool emitting, bool *changed) {
  KALDI_ASSERT(frame < active_toks_.size());
  Token *&toks = active_toks_[frame].toks;
    
  unordered_map<StateId, Token*>::iterator find_iter = cur_toks_.find(state);
  if (find_iter == cur_toks_.end()) { // no such token presently.
    // Create one.
    const BaseFloat extra_cost = 0.0;
    // tokens on the currently final frame have zero extra_cost
    // as any of them could end up
    // on the winning path.
    Token *new_tok = new Token (tot_cost, extra_cost, NULL, toks);
    toks = new_tok;
    num_toks_++;
    cur_toks_[state] = new_tok;
    if (changed) *changed = true;
    return new_tok;
  } else {
    Token *tok = find_iter->second; // There is an existing Token for this state.
    if (tok->tot_cost > tot_cost) {
      tok->tot_cost = tot_cost;
      if (changed) *changed = true;
    } else {
      if (changed) *changed = false;
    }
    return tok;
  }
}
  
// delta is the amount by which the extra_costs must
// change before it sets "extra_costs_changed" to true.  If delta is larger,
// we'll tend to go back less far toward the beginning of the file.
void LatticeSimpleDecoder::PruneForwardLinks(
    int32 frame, bool *extra_costs_changed,
    bool *links_pruned, BaseFloat delta) {
  // We have to iterate until there is no more change, because the links
  // are not guaranteed to be in topological order.

  *extra_costs_changed = false;
  *links_pruned = false;
  KALDI_ASSERT(frame >= 0 && frame < active_toks_.size());
  if (active_toks_[frame].toks == NULL ) { // empty list; this should
    // not happen.
    if (!warned_) {
      KALDI_WARN << "No tokens alive [doing pruning].. warning first "
          "time only for each utterance\n";
      warned_ = true;
    }
  }
    
  bool changed = true;
  while (changed) {
    changed = false;
    for (Token *tok = active_toks_[frame].toks; tok != NULL; tok = tok->next) {
      ForwardLink *link, *prev_link = NULL;
      // will recompute tok_extra_cost.
      BaseFloat tok_extra_cost = std::numeric_limits<BaseFloat>::infinity();
      for (link = tok->links; link != NULL; ) {
        // See if we need to excise this link...
        Token *next_tok = link->next_tok;
        BaseFloat link_extra_cost = next_tok->extra_cost +
            ((tok->tot_cost + link->acoustic_cost + link->graph_cost)
             - next_tok->tot_cost);
        KALDI_ASSERT(link_extra_cost == link_extra_cost); // check for NaN
        if (link_extra_cost > config_.lattice_beam) {  // excise link
          ForwardLink *next_link = link->next;
          if (prev_link != NULL) prev_link->next = next_link;
          else tok->links = next_link;
          delete link;
          link = next_link; // advance link but leave prev_link the same.
          *links_pruned = true;
        } else { // keep the link and update the tok_extra_cost if needed.
          if (link_extra_cost < 0.0) { // this is just a precaution.
            if (link_extra_cost < -0.01)
              KALDI_WARN << "Negative extra_cost: " << link_extra_cost;
            link_extra_cost = 0.0;
          }
          if (link_extra_cost < tok_extra_cost)
            tok_extra_cost = link_extra_cost;
          prev_link = link;
          link = link->next;
        }
      }
      if (fabs(tok_extra_cost - tok->extra_cost) > delta)
        changed = true;
      tok->extra_cost = tok_extra_cost; // will be +infinity or <= lattice_beam_.
    }
    if (changed) *extra_costs_changed = true;

    // Note: it's theoretically possible that aggressive compiler
    // optimizations could cause an infinite loop here for small delta and
    // high-dynamic-range scores.
  }
}


void LatticeSimpleDecoder::ComputeFinalCosts(
    unordered_map<Token*, BaseFloat> *final_costs,
    BaseFloat *final_relative_cost,
    BaseFloat *final_best_cost) const {
  KALDI_ASSERT(!decoding_finalized_);
  if (final_costs != NULL)
    final_costs->clear();

  BaseFloat infinity = std::numeric_limits<BaseFloat>::infinity();
  BaseFloat best_cost = infinity,
      best_cost_with_final = infinity;
  
  for (unordered_map<StateId, Token*>::const_iterator iter = cur_toks_.begin();
       iter != cur_toks_.end(); ++iter) {
    StateId state = iter->first;
    Token *tok = iter->second;
    BaseFloat final_cost = fst_.Final(state).Value();
    BaseFloat cost = tok->tot_cost,
        cost_with_final = cost + final_cost;
    best_cost = std::min(cost, best_cost);
    best_cost_with_final = std::min(cost_with_final, best_cost_with_final);
    if (final_costs != NULL && final_cost != infinity)
      (*final_costs)[tok] = final_cost;
  }
  if (final_relative_cost != NULL) {
    if (best_cost == infinity && best_cost_with_final == infinity) {
      // Likely this will only happen if there are no tokens surviving.
      // This seems the least bad way to handle it.
      *final_relative_cost = infinity;
    } else {
      *final_relative_cost = best_cost_with_final - best_cost;
    }
  }
  if (final_best_cost != NULL) {
    if (best_cost_with_final != infinity) { // final-state exists.
      *final_best_cost = best_cost_with_final;
    } else { // no final-state exists.
      *final_best_cost = best_cost;
    }
  }
}


// PruneForwardLinksFinal is a version of PruneForwardLinks that we call
// on the final frame.  If there are final tokens active, it uses the final-probs
// for pruning, otherwise it treats all tokens as final.
void LatticeSimpleDecoder::PruneForwardLinksFinal() {
  KALDI_ASSERT(!active_toks_.empty());  
  int32 frame_plus_one = active_toks_.size() - 1;

  if (active_toks_[frame_plus_one].toks == NULL) // empty list; should not happen.
    KALDI_WARN << "No tokens alive at end of file\n";

  typedef unordered_map<Token*, BaseFloat>::const_iterator IterType;  
  ComputeFinalCosts(&final_costs_, &final_relative_cost_, &final_best_cost_);
  decoding_finalized_ = true;
  // We're about to delete some of the tokens active on the final frame, so we
  // clear cur_toks_ because otherwise it would then contain dangling pointers.
  cur_toks_.clear();
  
  // Now go through tokens on this frame, pruning forward links...  may have to
  // iterate a few times until there is no more change, because the list is not
  // in topological order.  This is a modified version of the code in
  // PruneForwardLinks, but here we also take account of the final-probs.
  bool changed = true;
  BaseFloat delta = 1.0e-05;
  while (changed) {
    changed = false;
    for (Token *tok = active_toks_[frame_plus_one].toks;
         tok != NULL; tok = tok->next) {
      ForwardLink *link, *prev_link=NULL;
      // will recompute tok_extra_cost.  It has a term in it that corresponds
      // to the "final-prob", so instead of initializing tok_extra_cost to infinity
      // below we set it to the difference between the (score+final_prob) of this token,
      // and the best such (score+final_prob).

      BaseFloat final_cost;
      if (final_costs_.empty()) {
        final_cost = 0.0;
      } else {
        IterType iter = final_costs_.find(tok);
        if (iter != final_costs_.end())
          final_cost = iter->second;
        else
          final_cost = std::numeric_limits<BaseFloat>::infinity();
      }
      BaseFloat tok_extra_cost = tok->tot_cost + final_cost - final_best_cost_;
      // tok_extra_cost will be a "min" over either directly being final, or
      // being indirectly final through other links, and the loop below may
      // decrease its value:
      for (link = tok->links; link != NULL; ) {
        // See if we need to excise this link...
        Token *next_tok = link->next_tok;
        BaseFloat link_extra_cost = next_tok->extra_cost +
            ((tok->tot_cost + link->acoustic_cost + link->graph_cost)
             - next_tok->tot_cost);
        if (link_extra_cost > config_.lattice_beam) { // excise link
          ForwardLink *next_link = link->next;
          if (prev_link != NULL) prev_link->next = next_link;
          else tok->links = next_link;
          delete link;
          link = next_link; // advance link but leave prev_link the same.
        } else { // keep the link and update the tok_extra_cost if needed.
          if (link_extra_cost < 0.0) { // this is just a precaution.
            if (link_extra_cost < -0.01)
              KALDI_WARN << "Negative extra_cost: " << link_extra_cost;
            link_extra_cost = 0.0;
          }
          if (link_extra_cost < tok_extra_cost)
            tok_extra_cost = link_extra_cost;
          prev_link = link;
          link = link->next;
        }
      }
      // prune away tokens worse than lattice_beam above best path.  This step
      // was not necessary in the non-final case because then, this case
      // showed up as having no forward links.  Here, the tok_extra_cost has
      // an extra component relating to the final-prob.
      if (tok_extra_cost > config_.lattice_beam)
        tok_extra_cost = std::numeric_limits<BaseFloat>::infinity();
      // to be pruned in PruneTokensForFrame

      if (!ApproxEqual(tok->extra_cost, tok_extra_cost, delta))
        changed = true;
      tok->extra_cost = tok_extra_cost; // will be +infinity or <= lattice_beam_.
    }
  } // while changed
}

BaseFloat LatticeSimpleDecoder::FinalRelativeCost() const {
  if (!decoding_finalized_) {
    BaseFloat relative_cost;
    ComputeFinalCosts(NULL, &relative_cost, NULL);
    return relative_cost;
  } else {
    // we're not allowed to call that function if FinalizeDecoding() has
    // been called; return a cached value.
    return final_relative_cost_;
  }
}
  
// Prune away any tokens on this frame that have no forward links. [we don't do
// this in PruneForwardLinks because it would give us a problem with dangling
// pointers].
void LatticeSimpleDecoder::PruneTokensForFrame(int32 frame) {
  KALDI_ASSERT(frame >= 0 && frame < active_toks_.size());
  Token *&toks = active_toks_[frame].toks;
  if (toks == NULL)
    KALDI_WARN << "No tokens alive [doing pruning]";
  Token *tok, *next_tok, *prev_tok = NULL;
  for (tok = toks; tok != NULL; tok = next_tok) {
    next_tok = tok->next;
    if (tok->extra_cost == std::numeric_limits<BaseFloat>::infinity()) {
      // Next token is unreachable from end of graph; excise tok from list
      // and delete tok.
      if (prev_tok != NULL) prev_tok->next = tok->next;
      else toks = tok->next;
      delete tok;
      num_toks_--;
    } else {
      prev_tok = tok;
    }
  }
}
  
// Go backwards through still-alive tokens, pruning them, starting not from
// the current frame (where we want to keep all tokens) but from the frame before
// that.  We go backwards through the frames and stop when we reach a point
// where the delta-costs are not changing (and the delta controls when we consider
// a cost to have "not changed").
void LatticeSimpleDecoder::PruneActiveTokens(BaseFloat delta) {
  int32 cur_frame_plus_one = NumFramesDecoded();  
  int32 num_toks_begin = num_toks_;
  // The index "f" below represents a "frame plus one", i.e. you'd have to subtract
  // one to get the corresponding index for the decodable object.
  for (int32 f = cur_frame_plus_one - 1; f >= 0; f--) {
    // Reason why we need to prune forward links in this situation:
    // (1) we have never pruned them
    // (2) we never pruned the forward links on the next frame, which
    //     
    if (active_toks_[f].must_prune_forward_links) {
      bool extra_costs_changed = false, links_pruned = false;
      PruneForwardLinks(f, &extra_costs_changed, &links_pruned, delta);
      if (extra_costs_changed && f > 0)
        active_toks_[f-1].must_prune_forward_links = true;
      if (links_pruned)
        active_toks_[f].must_prune_tokens = true;
      active_toks_[f].must_prune_forward_links = false;
    }
    if (f+1 < cur_frame_plus_one && 
        active_toks_[f+1].must_prune_tokens) {
      PruneTokensForFrame(f+1);
      active_toks_[f+1].must_prune_tokens = false;
    }
  }
  KALDI_VLOG(3) << "PruneActiveTokens: pruned tokens from " << num_toks_begin
                << " to " << num_toks_;
}


// FinalizeDecoding() is a version of PruneActiveTokens that we call
// (optionally) on the final frame.  Takes into account the final-prob of
// tokens.  This function used to be called PruneActiveTokensFinal().
void LatticeSimpleDecoder::FinalizeDecoding() {
  int32 final_frame_plus_one = NumFramesDecoded();  
  int32 num_toks_begin = num_toks_;
  PruneForwardLinksFinal();
  for (int32 f = final_frame_plus_one - 1; f >= 0; f--) {  
    bool b1, b2; // values not used.
    BaseFloat dontcare = 0.0;
    PruneForwardLinks(f, &b1, &b2, dontcare);
    PruneTokensForFrame(f + 1);
  }
  PruneTokensForFrame(0); 
  KALDI_VLOG(3) << "pruned tokens from " << num_toks_begin
                << " to " << num_toks_;
}
  
void LatticeSimpleDecoder::ProcessEmitting(DecodableInterface *decodable) {
  int32 frame = active_toks_.size() - 1; // frame is the frame-index
                                         // (zero-based) used to get likelihoods
                                         // from the decodable object.
  active_toks_.resize(active_toks_.size() + 1);
  prev_toks_.clear();
  cur_toks_.swap(prev_toks_);

  // Processes emitting arcs for one frame.  Propagates from
  // prev_toks_ to cur_toks_.
  BaseFloat cutoff = std::numeric_limits<BaseFloat>::infinity();
  for (unordered_map<StateId, Token*>::iterator iter = prev_toks_.begin();
       iter != prev_toks_.end();
       ++iter) {
    StateId state = iter->first;
    Token *tok = iter->second;
    for (fst::ArcIterator<fst::Fst<Arc> > aiter(fst_, state);
         !aiter.Done();
         aiter.Next()) {
      const Arc &arc = aiter.Value();
      if (arc.ilabel != 0) {  // propagate..
        BaseFloat ac_cost = -decodable->LogLikelihood(frame, arc.ilabel),
            graph_cost = arc.weight.Value(),
            cur_cost = tok->tot_cost,
            tot_cost = cur_cost + ac_cost + graph_cost;
        if (tot_cost >= cutoff) continue;
        else if (tot_cost + config_.beam < cutoff)
          cutoff = tot_cost + config_.beam;
        // AddToken adds the next_tok to cur_toks_ (if not already present).
        Token *next_tok = FindOrAddToken(arc.nextstate, frame + 1, tot_cost,
                                         true, NULL);
          
        // Add ForwardLink from tok to next_tok (put on head of list tok->links)
        tok->links = new ForwardLink(next_tok, arc.ilabel, arc.olabel, 
                                     graph_cost, ac_cost, tok->links);
      }
    }
  }
}

void LatticeSimpleDecoder::ProcessNonemitting() {
  KALDI_ASSERT(!active_toks_.empty());
  int32 frame = static_cast<int32>(active_toks_.size()) - 2;
  // Note: "frame" is the time-index we just processed, or -1 if
  // we are processing the nonemitting transitions before the
  // first frame (called from InitDecoding()).
    
  // Processes nonemitting arcs for one frame.  Propagates within
  // cur_toks_.  Note-- this queue structure is is not very optimal as
  // it may cause us to process states unnecessarily (e.g. more than once),
  // but in the baseline code, turning this vector into a set to fix this
  // problem did not improve overall speed.
  std::vector<StateId> queue;
  BaseFloat best_cost = std::numeric_limits<BaseFloat>::infinity();
  for (unordered_map<StateId, Token*>::iterator iter = cur_toks_.begin();
       iter != cur_toks_.end();
       ++iter) {
    StateId state = iter->first;
    if (fst_.NumInputEpsilons(state) != 0)
      queue.push_back(state);
    best_cost = std::min(best_cost, iter->second->tot_cost);
  }
  if (queue.empty()) {
    if (!warned_) {
      KALDI_ERR << "Error in ProcessEmitting: no surviving tokens: frame is "
                << frame;
      warned_ = true;
    }
  }
  BaseFloat cutoff = best_cost + config_.beam;
    
  while (!queue.empty()) {
    StateId state = queue.back();
    queue.pop_back();
    Token *tok = cur_toks_[state];
    // If "tok" has any existing forward links, delete them,
    // because we're about to regenerate them.  This is a kind
    // of non-optimality (remember, this is the simple decoder),
    // but since most states are emitting it's not a huge issue.
    tok->DeleteForwardLinks();
    tok->links = NULL;
    for (fst::ArcIterator<fst::Fst<Arc> > aiter(fst_, state);
         !aiter.Done();
         aiter.Next()) {
      const Arc &arc = aiter.Value();
      if (arc.ilabel == 0) {  // propagate nonemitting only...
        BaseFloat graph_cost = arc.weight.Value(),
            cur_cost = tok->tot_cost,
            tot_cost = cur_cost + graph_cost;
        if (tot_cost < cutoff) {
          bool changed;
          Token *new_tok = FindOrAddToken(arc.nextstate, frame + 1, tot_cost,
                                          false, &changed);
          
          tok->links = new ForwardLink(new_tok, 0, arc.olabel,
                                       graph_cost, 0, tok->links);
            
          // "changed" tells us whether the new token has a different
          // cost from before, or is new [if so, add into queue].
          if (changed && fst_.NumInputEpsilons(arc.nextstate) != 0)
            queue.push_back(arc.nextstate);
        }
      }
    }
  }
}

void LatticeSimpleDecoder::ClearActiveTokens() { // a cleanup routine, at utt end/begin
  for (size_t i = 0; i < active_toks_.size(); i++) {
    // Delete all tokens alive on this frame, and any forward
    // links they may have.
    for (Token *tok = active_toks_[i].toks; tok != NULL; ) {
      tok->DeleteForwardLinks();
      Token *next_tok = tok->next;
      delete tok;
      num_toks_--;
      tok = next_tok;
    }
  }
  active_toks_.clear();
  KALDI_ASSERT(num_toks_ == 0);
}

// PruneCurrentTokens deletes the tokens from the "toks" map, but not
// from the active_toks_ list, which could cause dangling forward pointers
// (will delete it during regular pruning operation).
void LatticeSimpleDecoder::PruneCurrentTokens(BaseFloat beam, unordered_map<StateId, Token*> *toks) {
  if (toks->empty()) {
    KALDI_VLOG(2) <<  "No tokens to prune.\n";
    return;
  }
  BaseFloat best_cost = 1.0e+10;  // positive == high cost == bad.
  for (unordered_map<StateId, Token*>::iterator iter = toks->begin();
       iter != toks->end(); ++iter) {
    best_cost =
        std::min(best_cost,
                 static_cast<BaseFloat>(iter->second->tot_cost));
  }
  std::vector<StateId> retained;
  BaseFloat cutoff = best_cost + beam;
  for (unordered_map<StateId, Token*>::iterator iter = toks->begin();
       iter != toks->end(); ++iter) {
    if (iter->second->tot_cost < cutoff)
      retained.push_back(iter->first);
  }
  unordered_map<StateId, Token*> tmp;
  for (size_t i = 0; i < retained.size(); i++) {
    tmp[retained[i]] = (*toks)[retained[i]];
  }
  KALDI_VLOG(2) <<  "Pruned to "<<(retained.size())<<" toks.\n";
  tmp.swap(*toks);
}


} // end namespace kaldi.