WordPruningBreadthFirstSearchManager (sphinx4-core 5prealpha-SNAPSHOT API)

java.lang.Object
- edu.cmu.sphinx.decoder.search.TokenSearchManager
- - edu.cmu.sphinx.decoder.search.WordPruningBreadthFirstSearchManager

All Implemented Interfaces:

SearchManager, Configurable

Direct Known Subclasses:

TokenHeapSearchManager, WordPruningBreadthFirstLookaheadSearchManager
```
public class WordPruningBreadthFirstSearchManager
extends TokenSearchManager
```
Provides the breadth first search. To perform recognition an application should call initialize before recognition begins, and repeatedly call recognize until Result.isFinal() returns true. Once a final result has been obtained, stopRecognition should be called.
All scores and probabilities are maintained in the log math log domain.

Field Summary

Fields
Modifier and Type	Field and Description
`protected float`	`acousticLookaheadFrames`
`protected ActiveList`	`activeList`
`protected java.util.Map<SearchState,Token>`	`bestTokenMap`
`protected boolean`	`checkStateOrder`
`protected long`	`currentCollectTime`
`protected int`	`currentFrameNumber`
`protected StatisticsVariable`	`curTokensScored`
`protected Timer`	`growTimer`
`protected Linguist`	`linguist`
`protected java.util.logging.Logger`	`logger`
`protected LogMath`	`logMath`
`protected AlternateHypothesisManager`	`loserManager`
`static java.lang.String`	`PROP_ACOUSTIC_LOOKAHEAD_FRAMES` The property that controls the amount of simple acoustic lookahead performed.
`static java.lang.String`	`PROP_ACTIVE_LIST_MANAGER` The property that defines the type of active list to use
`static java.lang.String`	`PROP_CHECK_STATE_ORDER` The property for checking if the order of states is valid.
`static java.lang.String`	`PROP_GROW_SKIP_INTERVAL` The property that controls the number of frames processed for every time the decode growth step is skipped.
`static java.lang.String`	`PROP_LINGUIST` The property that defines the name of the linguist to be used by this search manager.
`static java.lang.String`	`PROP_MAX_LATTICE_EDGES` The property that specifies the maximum lattice edges
`static java.lang.String`	`PROP_PRUNER` The property that defines the name of the linguist to be used by this search manager.
`static java.lang.String`	`PROP_RELATIVE_BEAM_WIDTH` The property that specifies the relative beam width
`static java.lang.String`	`PROP_SCORER` The property that defines the name of the scorer to be used by this search manager.
`static java.lang.String`	`PROP_SHOW_TOKEN_COUNT` The property than, when set to `true` will cause the recognizer to count up all the tokens in the active list after every frame.
`protected Pruner`	`pruner`
`protected Timer`	`pruneTimer`
`protected float`	`relativeBeamWidth`
`protected java.util.List<Token>`	`resultList`
`protected AcousticScorer`	`scorer`
`protected Timer`	`scoreTimer`
`protected boolean`	`showTokenCount`
`protected boolean`	`streamEnd`
`protected StatisticsVariable`	`tokensCreated`
`protected StatisticsVariable`	`totalTokensScored`

Fields inherited from class edu.cmu.sphinx.decoder.search.TokenSearchManager
buildWordLattice, keepAllTokens, PROP_BUILD_WORD_LATTICE, PROP_KEEP_ALL_TOKENS

Constructor Summary

Constructors
Constructor and Description
`WordPruningBreadthFirstSearchManager()`
`WordPruningBreadthFirstSearchManager(Linguist linguist, Pruner pruner, AcousticScorer scorer, ActiveListManager activeListManager, boolean showTokenCount, double relativeWordBeamWidth, int growSkipInterval, boolean checkStateOrder, boolean buildWordLattice, int maxLatticeEdges, float acousticLookaheadFrames, boolean keepAllTokens)` Creates a pruning manager withs separate lists for tokens

Method Summary

All Methods Instance Methods Concrete Methods
Modifier and Type	Method and Description
`protected void`	`activeListAdd(Token token)`
`void`	`allocate()` Allocates the resources necessary for this search.
`protected boolean`	`allowExpansion(Token t)` Determine if the given token should be expanded
`protected void`	`checkStateOrder(SearchState fromState, SearchState toState)` Checks that the given two states are in legitimate order.
`protected void`	`collectSuccessorTokens(Token token)` Collects the next set of emitting tokens from a token and accumulates them in the active or result lists
`protected void`	`createBestTokenMap()` creates a new best token map with the best size
`void`	`deallocate()` Deallocates resources necessary for this search.
`ActiveList`	`getActiveList()` Returns the ActiveList.
`protected Token`	`getBestToken(SearchState state)` Gets the best token for this state
`int`	`getCurrentFrameNumber()` Returns the current frame number.
`Timer`	`getGrowTimer()` Returns the Timer for growing.
`java.util.List<Token>`	`getResultList()` Returns the result list.
`StatisticsVariable`	`getTokensCreated()` Returns the tokensCreated StatisticsVariable.
`protected void`	`growBranches()` Goes through the active list of tokens and expands each token, finding the set of successor tokens until all the successor tokens are emitting tokens.
`protected void`	`growEmittingBranches()` Grows the emitting branches.
`protected boolean`	`isVisited(Token t)` Determines whether or not we've visited the state associated with this token since the previous frame.
`protected void`	`localStart()` Gets the initial grammar node from the linguist and creates a GrammarNodeToken
`protected void`	`localStop()` Local cleanup for this search manager
`protected void`	`monitorStates(ActiveList activeList)` Keeps track of and reports statistics about the number of active states
`void`	`newProperties(PropertySheet ps)` This method is called when this configurable component needs to be reconfigured.
`protected void`	`pruneBranches()` Removes unpromising branches from the active list
`protected boolean`	`recognize()`
`Result`	`recognize(int nFrames)` Performs the recognition for the given number of frames.
`protected boolean`	`scoreTokens()` Calculate the acoustic scores for the active list.
`void`	`setActiveList(ActiveList activeList)` Sets the ActiveList.
`protected void`	`setBestToken(Token token, SearchState state)` Sets the best token for a given state
`void`	`setResultList(java.util.List<Token> resultList)` Sets the result list.
`protected void`	`showTokenCount()` Counts all the tokens in the active list (and displays them).
`void`	`startRecognition()` Called at the start of recognition.
`void`	`stopRecognition()` Terminates a recognition

Methods inherited from class edu.cmu.sphinx.decoder.search.TokenSearchManager
getResultListPredecessor

Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

- Field Detail
  - PROP_LINGUIST
```
@S4Component(type=Linguist.class)
public static final java.lang.String PROP_LINGUIST
```
    The property that defines the name of the linguist to be used by this search manager.
    
    See Also:
    
    Constant Field Values
  - PROP_PRUNER
```
@S4Component(type=Pruner.class)
public static final java.lang.String PROP_PRUNER
```
    The property that defines the name of the linguist to be used by this search manager.
    
    See Also:
    
    Constant Field Values
  - PROP_SCORER
```
@S4Component(type=AcousticScorer.class)
public static final java.lang.String PROP_SCORER
```
    The property that defines the name of the scorer to be used by this search manager.
    
    See Also:
    
    Constant Field Values
  - PROP_SHOW_TOKEN_COUNT
```
@S4Boolean(defaultValue=false)
public static final java.lang.String PROP_SHOW_TOKEN_COUNT
```
    The property than, when set to true will cause the recognizer to count up all the tokens in the active list after every frame.
    
    See Also:
    
    Constant Field Values
  - PROP_GROW_SKIP_INTERVAL
```
@S4Integer(defaultValue=0)
public static final java.lang.String PROP_GROW_SKIP_INTERVAL
```
    The property that controls the number of frames processed for every time the decode growth step is skipped. Setting this property to zero disables grow skipping. Setting this number to a small integer will increase the speed of the decoder but will also decrease its accuracy. The higher the number, the less often the grow code is skipped. Values like 6-8 is known to be the good enough for large vocabulary tasks. That means that one of 6 frames will be skipped.
    
    See Also:
    
    Constant Field Values
  - PROP_ACTIVE_LIST_MANAGER
```
@S4Component(type=ActiveListManager.class)
public static final java.lang.String PROP_ACTIVE_LIST_MANAGER
```
    The property that defines the type of active list to use
    
    See Also:
    
    Constant Field Values
  - PROP_CHECK_STATE_ORDER
```
@S4Boolean(defaultValue=false)
public static final java.lang.String PROP_CHECK_STATE_ORDER
```
    The property for checking if the order of states is valid.
    
    See Also:
    
    Constant Field Values
  - PROP_MAX_LATTICE_EDGES
```
@S4Integer(defaultValue=100)
public static final java.lang.String PROP_MAX_LATTICE_EDGES
```
    The property that specifies the maximum lattice edges
    
    See Also:
    
    Constant Field Values
  - PROP_ACOUSTIC_LOOKAHEAD_FRAMES
```
@S4Double(defaultValue=0.0)
public static final java.lang.String PROP_ACOUSTIC_LOOKAHEAD_FRAMES
```
    The property that controls the amount of simple acoustic lookahead performed. Setting the property to zero (the default) disables simple acoustic lookahead. The lookahead need not be an integer.
    
    See Also:
    
    Constant Field Values
  - PROP_RELATIVE_BEAM_WIDTH
```
@S4Double(defaultValue=0.0)
public static final java.lang.String PROP_RELATIVE_BEAM_WIDTH
```
    The property that specifies the relative beam width
    
    See Also:
    
    Constant Field Values
  - linguist
```
protected Linguist linguist
```
  - pruner
```
protected Pruner pruner
```
  - scorer
```
protected AcousticScorer scorer
```
  - logMath
```
protected LogMath logMath
```
  - logger
```
protected java.util.logging.Logger logger
```
  - showTokenCount
```
protected boolean showTokenCount
```
  - checkStateOrder
```
protected boolean checkStateOrder
```
  - relativeBeamWidth
```
protected float relativeBeamWidth
```
  - acousticLookaheadFrames
```
protected float acousticLookaheadFrames
```
  - scoreTimer
```
protected Timer scoreTimer
```
  - pruneTimer
```
protected Timer pruneTimer
```
  - growTimer
```
protected Timer growTimer
```
  - totalTokensScored
```
protected StatisticsVariable totalTokensScored
```
  - curTokensScored
```
protected StatisticsVariable curTokensScored
```
  - tokensCreated
```
protected StatisticsVariable tokensCreated
```
  - currentFrameNumber
```
protected int currentFrameNumber
```
  - currentCollectTime
```
protected long currentCollectTime
```
  - activeList
```
protected ActiveList activeList
```
  - resultList
```
protected java.util.List<Token> resultList
```
  - bestTokenMap
```
protected java.util.Map<SearchState,Token> bestTokenMap
```
  - loserManager
```
protected AlternateHypothesisManager loserManager
```
  - streamEnd
```
protected boolean streamEnd
```
- Constructor Detail
  - WordPruningBreadthFirstSearchManager
```
public WordPruningBreadthFirstSearchManager(Linguist linguist,
                                            Pruner pruner,
                                            AcousticScorer scorer,
                                            ActiveListManager activeListManager,
                                            boolean showTokenCount,
                                            double relativeWordBeamWidth,
                                            int growSkipInterval,
                                            boolean checkStateOrder,
                                            boolean buildWordLattice,
                                            int maxLatticeEdges,
                                            float acousticLookaheadFrames,
                                            boolean keepAllTokens)
```
    Creates a pruning manager withs separate lists for tokens
    
    Parameters:
    
    linguist - a linguist for search space
    
    pruner - pruner to drop tokens
    
    scorer - scorer to estimate token probability
    
    activeListManager - active list manager to store tokens
    
    showTokenCount - show count during decoding
    
    relativeWordBeamWidth - relative beam for lookahead pruning
    
    growSkipInterval - skip interval for grown
    
    checkStateOrder - check order of states during growth
    
    buildWordLattice - build a lattice during decoding
    
    maxLatticeEdges - max edges to keep in lattice
    
    acousticLookaheadFrames - frames to do lookahead
    
    keepAllTokens - keep tokens including emitting tokens
  - WordPruningBreadthFirstSearchManager
```
public WordPruningBreadthFirstSearchManager()
```
- Method Detail
  - newProperties
```
public void newProperties(PropertySheet ps)
                   throws PropertyException
```
    Description copied from interface: Configurable
    
    This method is called when this configurable component needs to be reconfigured.
    
    Specified by:
    
    newProperties in interface Configurable
    
    Overrides:
    
    newProperties in class TokenSearchManager
    
    Parameters:
    
    ps - a property sheet holding the new data
    
    Throws:
    
    PropertyException - if there is a problem with the properties.
  - allocate
```
public void allocate()
```
    Description copied from interface: SearchManager
    
    Allocates the resources necessary for this search. This should be called once before an recognitions are performed
  - deallocate
```
public void deallocate()
```
    Description copied from interface: SearchManager
    
    Deallocates resources necessary for this search. This should be called once after all recognitions are completed at the search manager is no longer needed.
  - startRecognition
```
public void startRecognition()
```
    Called at the start of recognition. Gets the search manager ready to recognize
  - recognize
```
public Result recognize(int nFrames)
```
    Performs the recognition for the given number of frames.
    
    Parameters:
    
    nFrames - the number of frames to recognize
    
    Returns:
    
    the current result
  - recognize
```
protected boolean recognize()
```
  - createBestTokenMap
```
protected void createBestTokenMap()
```
    creates a new best token map with the best size
  - stopRecognition
```
public void stopRecognition()
```
    Terminates a recognition
  - localStart
```
protected void localStart()
```
    Gets the initial grammar node from the linguist and creates a GrammarNodeToken
  - localStop
```
protected void localStop()
```
    Local cleanup for this search manager
  - growBranches
```
protected void growBranches()
```
    Goes through the active list of tokens and expands each token, finding the set of successor tokens until all the successor tokens are emitting tokens.
  - growEmittingBranches
```
protected void growEmittingBranches()
```
    Grows the emitting branches. This version applies a simple acoustic lookahead based upon the rate of change in the current acoustic score.
  - scoreTokens
```
protected boolean scoreTokens()
```
    Calculate the acoustic scores for the active list. The active list should contain only emitting tokens.
    
    Returns:
    
    true if there are more frames to score, otherwise, false
  - monitorStates
```
protected void monitorStates(ActiveList activeList)
```
    Keeps track of and reports statistics about the number of active states
    
    Parameters:
    
    activeList - the active list of states
  - pruneBranches
```
protected void pruneBranches()
```
    Removes unpromising branches from the active list
  - getBestToken
```
protected Token getBestToken(SearchState state)
```
    Gets the best token for this state
    
    Parameters:
    
    state - the state of interest
    
    Returns:
    
    the best token
  - setBestToken
```
protected void setBestToken(Token token,
                            SearchState state)
```
    Sets the best token for a given state
    
    Parameters:
    
    token - the best token
    
    state - the state
  - checkStateOrder
```
protected void checkStateOrder(SearchState fromState,
                               SearchState toState)
```
    Checks that the given two states are in legitimate order.
    
    Parameters:
    
    fromState - parent state
    
    toState - child state
  - collectSuccessorTokens
```
protected void collectSuccessorTokens(Token token)
```
    Collects the next set of emitting tokens from a token and accumulates them in the active or result lists
    
    Parameters:
    
    token - the token to collect successors from be immediately expanded are placed. Null if we should always expand all nodes.
  - isVisited
```
protected boolean isVisited(Token t)
```
    Determines whether or not we've visited the state associated with this token since the previous frame.
    
    Parameters:
    
    t - token to check
    
    Returns:
    
    true if we've visited the search state since the last frame
  - activeListAdd
```
protected void activeListAdd(Token token)
```
  - allowExpansion
```
protected boolean allowExpansion(Token t)
```
    Determine if the given token should be expanded
    
    Parameters:
    
    t - the token to test
    
    Returns:
    
    true if the token should be expanded
  - showTokenCount
```
protected void showTokenCount()
```
    Counts all the tokens in the active list (and displays them). This is an expensive operation.
  - getActiveList
```
public ActiveList getActiveList()
```
    Returns the ActiveList.
    
    Returns:
    
    the ActiveList
  - setActiveList
```
public void setActiveList(ActiveList activeList)
```
    Sets the ActiveList.
    
    Parameters:
    
    activeList - the new ActiveList
  - getResultList
```
public java.util.List<Token> getResultList()
```
    Returns the result list.
    
    Returns:
    
    the result list
  - setResultList
```
public void setResultList(java.util.List<Token> resultList)
```
    Sets the result list.
    
    Parameters:
    
    resultList - the new result list
  - getCurrentFrameNumber
```
public int getCurrentFrameNumber()
```
    Returns the current frame number.
    
    Returns:
    
    the current frame number
  - getGrowTimer
```
public Timer getGrowTimer()
```
    Returns the Timer for growing.
    
    Returns:
    
    the Timer for growing
  - getTokensCreated
```
public StatisticsVariable getTokensCreated()
```
    Returns the tokensCreated StatisticsVariable.
    
    Returns:
    
    the tokensCreated StatisticsVariable.

Class WordPruningBreadthFirstSearchManager

Field Summary

Fields inherited from class edu.cmu.sphinx.decoder.search.TokenSearchManager

Constructor Summary

Method Summary

Methods inherited from class edu.cmu.sphinx.decoder.search.TokenSearchManager

Methods inherited from class java.lang.Object

Field Detail

PROP_LINGUIST

PROP_PRUNER

PROP_SCORER

PROP_SHOW_TOKEN_COUNT

PROP_GROW_SKIP_INTERVAL

PROP_ACTIVE_LIST_MANAGER

PROP_CHECK_STATE_ORDER

PROP_MAX_LATTICE_EDGES

PROP_ACOUSTIC_LOOKAHEAD_FRAMES

PROP_RELATIVE_BEAM_WIDTH

linguist

pruner

scorer

logMath

logger

showTokenCount

checkStateOrder

relativeBeamWidth

acousticLookaheadFrames

scoreTimer

pruneTimer

growTimer

totalTokensScored

curTokensScored

tokensCreated

currentFrameNumber

currentCollectTime

activeList

resultList

bestTokenMap

loserManager

streamEnd

Constructor Detail

WordPruningBreadthFirstSearchManager

WordPruningBreadthFirstSearchManager

Method Detail

newProperties

allocate

deallocate

startRecognition

recognize

recognize

createBestTokenMap

stopRecognition

localStart

localStop

growBranches

growEmittingBranches

scoreTokens

monitorStates

pruneBranches

getBestToken

setBestToken

checkStateOrder

collectSuccessorTokens

isVisited

activeListAdd

allowExpansion

showTokenCount

getActiveList

setActiveList

getResultList

setResultList

getCurrentFrameNumber

getGrowTimer

getTokensCreated