edu.cmu.sphinx.trainer

Interface Learner

All Superinterfaces:

Configurable

All Known Implementing Classes:

BaumWelchLearner, FlatInitializerLearner

public interface Learner
extends Configurable

Provides mechanisms for computing statistics given a set of states and input data.

Method Summary

Modifier and Type	Method and Description
TrainerScore[]	getScore()
void	initializeComputation(Utterance utterance, UtteranceGraph graph) Initializes computation for current utterance and utterance graph.
void	setGraph(UtteranceGraph graph) Implements the setGraph method.
void	setUtterance(Utterance utterance) Sets the learner to use a utterance.
void	start() Starts the Learner.
void	stop() Stops the Learner.

Methods inherited from interface edu.cmu.sphinx.util.props.Configurable

newProperties

Method Detail

start

void start()

Starts the Learner.

stop

void stop()

Stops the Learner.

setUtterance

void setUtterance(Utterance utterance)
           throws java.io.IOException

Sets the learner to use a utterance.

Parameters:

utterance - the utterance

Throws:

java.io.IOException - if exception occurred

initializeComputation

void initializeComputation(Utterance utterance,
                           UtteranceGraph graph)
                    throws java.io.IOException

Initializes computation for current utterance and utterance graph.

Parameters:

utterance - the current utterance

graph - the current utterance graph

Throws:

java.io.IOException - if exception occurred

setGraph

void setGraph(UtteranceGraph graph)

Implements the setGraph method. Since the flat initializer does not need a graph, this method produces an error.

Parameters:

graph - the graph

getScore

TrainerScore[] getScore()

Returns:

posterior probabilities for a given state.