You will instantiate this class when you want to decode a single utterance using the online-decoding setup for neural nets. More...

#include <online-nnet3-incremental-decoding.h>

Collaboration diagram for SingleUtteranceNnet3IncrementalDecoderTpl< FST >:

[legend]

Public Member Functions
	SingleUtteranceNnet3IncrementalDecoderTpl (const LatticeIncrementalDecoderConfig &decoder_opts, const TransitionModel &trans_model, const nnet3::DecodableNnetSimpleLoopedInfo &info, const FST &fst, OnlineNnet2FeaturePipeline *features)

void	InitDecoding (int32 frame_offset=0)
	Initializes the decoding and sets the frame offset of the underlying decodable object. More...

void	AdvanceDecoding ()
	Advances the decoding as far as we can. More...

void	FinalizeDecoding ()
	Finalizes the decoding. More...

int32	NumFramesDecoded () const

int32	NumFramesInLattice () const

const CompactLattice &	GetLattice (int32 num_frames_to_include, bool use_final_probs=false)

void	GetBestPath (bool end_of_utterance, Lattice *best_path) const
	Outputs an FST corresponding to the single best path through the current lattice. More...

bool	EndpointDetected (const OnlineEndpointConfig &config)
	This function calls EndpointDetected from online-endpoint.h, with the required arguments. More...

const LatticeIncrementalOnlineDecoderTpl< FST > &	Decoder () const

	~SingleUtteranceNnet3IncrementalDecoderTpl ()

Private Attributes
const LatticeIncrementalDecoderConfig &	decoder_opts_

BaseFloat	input_feature_frame_shift_in_seconds_

const TransitionModel &	trans_model_

nnet3::DecodableAmNnetLoopedOnline	decodable_

LatticeIncrementalOnlineDecoderTpl< FST >	decoder_

Detailed Description

template<typename FST>
class kaldi::SingleUtteranceNnet3IncrementalDecoderTpl< FST >

You will instantiate this class when you want to decode a single utterance using the online-decoding setup for neural nets.

The template will be instantiated only for FST = fst::Fst<fst::StdArc> and FST = fst::GrammarFst.

Definition at line 51 of file online-nnet3-incremental-decoding.h.

Constructor & Destructor Documentation

◆ SingleUtteranceNnet3IncrementalDecoderTpl()

SingleUtteranceNnet3IncrementalDecoderTpl	(	const LatticeIncrementalDecoderConfig &	decoder_opts,
		const TransitionModel &	trans_model,
		const nnet3::DecodableNnetSimpleLoopedInfo &	info,
		const FST &	fst,
		OnlineNnet2FeaturePipeline *	features
	)

Definition at line 28 of file online-nnet3-incremental-decoding.cc.

References SingleUtteranceNnet3IncrementalDecoderTpl< FST >::decoder_.

                                          :
     decoder_opts_(decoder_opts),
     input_feature_frame_shift_in_seconds_(features->FrameShiftInSeconds()),
     trans_model_(trans_model),
     decodable_(trans_model_, info,
                features->InputFeature(), features->IvectorFeature()),
     decoder_(fst, trans_model, decoder_opts_) {
   decoder_.InitDecoding();
 }

◆ ~SingleUtteranceNnet3IncrementalDecoderTpl()

~SingleUtteranceNnet3IncrementalDecoderTpl ( )

inline

Definition at line 120 of file online-nnet3-incremental-decoding.h.

120 { }

Member Function Documentation

◆ AdvanceDecoding()

void AdvanceDecoding ( )

Advances the decoding as far as we can.

Definition at line 50 of file online-nnet3-incremental-decoding.cc.

References SingleUtteranceNnet3IncrementalDecoderTpl< FST >::decodable_, and SingleUtteranceNnet3IncrementalDecoderTpl< FST >::decoder_.

Referenced by main().

                                                                      {
   decoder_.AdvanceDecoding(&decodable_);
 }

◆ Decoder()

const LatticeIncrementalOnlineDecoderTpl<FST>& Decoder ( ) const

inline

Definition at line 118 of file online-nnet3-incremental-decoding.h.

References SingleUtteranceNnet3IncrementalDecoderTpl< FST >::decoder_.

Referenced by main().

118 { return decoder_; }

kaldi::SingleUtteranceNnet3IncrementalDecoderTpl::decoder_

LatticeIncrementalOnlineDecoderTpl< FST > decoder_

Definition: online-nnet3-incremental-decoding.h:135

◆ EndpointDetected()

bool EndpointDetected ( const OnlineEndpointConfig & config )

This function calls EndpointDetected from online-endpoint.h, with the required arguments.

Definition at line 61 of file online-nnet3-incremental-decoding.cc.

References SingleUtteranceNnet3IncrementalDecoderTpl< FST >::decodable_, SingleUtteranceNnet3IncrementalDecoderTpl< FST >::decoder_, kaldi::EndpointDetected(), DecodableNnetLoopedOnlineBase::FrameSubsamplingFactor(), SingleUtteranceNnet3IncrementalDecoderTpl< FST >::input_feature_frame_shift_in_seconds_, and SingleUtteranceNnet3IncrementalDecoderTpl< FST >::trans_model_.

Referenced by SingleUtteranceNnet3IncrementalDecoderTpl< FST >::GetLattice(), and main().

                                         {
   BaseFloat output_frame_shift =
       input_feature_frame_shift_in_seconds_ *
       decodable_.FrameSubsamplingFactor();
   return kaldi::EndpointDetected(config, trans_model_,
                                  output_frame_shift, decoder_);
 }

◆ FinalizeDecoding()

void FinalizeDecoding ( )

inline

Finalizes the decoding.

Cleans up and prunes remaining tokens, so the GetLattice() call will return faster. You must not call this before calling (TerminateDecoding() or InputIsFinished()) and then Wait().

Definition at line 74 of file online-nnet3-incremental-decoding.h.

References SingleUtteranceNnet3IncrementalDecoderTpl< FST >::decoder_.

Referenced by main().

74 { decoder_.FinalizeDecoding(); }

kaldi::SingleUtteranceNnet3IncrementalDecoderTpl::decoder_

LatticeIncrementalOnlineDecoderTpl< FST > decoder_

Definition: online-nnet3-incremental-decoding.h:135

◆ GetBestPath()

void GetBestPath	(	bool	end_of_utterance,
		Lattice *	best_path
	)		const

Outputs an FST corresponding to the single best path through the current lattice.

If "use_final_probs" is true AND we reached the final-state of the graph then it will include those as final-probs, else it will treat all final-probs as one.

Definition at line 55 of file online-nnet3-incremental-decoding.cc.

References SingleUtteranceNnet3IncrementalDecoderTpl< FST >::decoder_.

Referenced by SingleUtteranceNnet3IncrementalDecoderTpl< FST >::GetLattice().

                                                                         {
   decoder_.GetBestPath(best_path, end_of_utterance);
 }

◆ GetLattice()

const CompactLattice& GetLattice	(	int32	num_frames_to_include,
		bool	use_final_probs = `false`
	)

inline

Definition at line 97 of file online-nnet3-incremental-decoding.h.

References SingleUtteranceNnet3IncrementalDecoderTpl< FST >::decoder_, SingleUtteranceNnet3IncrementalDecoderTpl< FST >::EndpointDetected(), and SingleUtteranceNnet3IncrementalDecoderTpl< FST >::GetBestPath().

Referenced by main().

                                                                     {
     return decoder_.GetLattice(num_frames_to_include, use_final_probs);
   }

◆ InitDecoding()

void InitDecoding ( int32 frame_offset = 0 )

Initializes the decoding and sets the frame offset of the underlying decodable object.

This method is called by the constructor. You can also call this method when you want to reset the decoder state, but want to keep using the same decodable object, e.g. in case of an endpoint.

Definition at line 44 of file online-nnet3-incremental-decoding.cc.

References SingleUtteranceNnet3IncrementalDecoderTpl< FST >::decodable_, SingleUtteranceNnet3IncrementalDecoderTpl< FST >::decoder_, and DecodableNnetLoopedOnlineBase::SetFrameOffset().

                                                                                     {
   decoder_.InitDecoding();
   decodable_.SetFrameOffset(frame_offset);
 }